def logic(self, image): for param in self.augmentation_params: self.augmentation_data.append([ str(param.augmentation_value), iaa.Salt(p=param.augmentation_value).to_deterministic(). augment_image(image), param.detection_tag ])
def __init__(self, batch_size, input_shape, anchors, num_classes, real_annotation_lines, is_training=True): self.batch_size = batch_size self.input_shape = input_shape self.anchors = anchors self.num_classes = num_classes self.real_annotation_lines = real_annotation_lines self.is_training = is_training sometimes = lambda aug: iaa.Sometimes(0.5, aug) self.aug_pipe = iaa.Sequential([ iaa.SomeOf((0, 5), [ iaa.Sharpen(alpha=1.0, lightness=(0.75, 1.5)), iaa.EdgeDetect(alpha=(0, 0.5)), iaa.AdditiveGaussianNoise(loc=0, scale=(0.0, 0.05 * 255)), iaa.OneOf([ iaa.Dropout((0.01, 0.05)), iaa.Salt((0.03, 0.15)), ]), iaa.Add((-10, 10)), iaa.Multiply((0.5, 1.5)), iaa.ContrastNormalization((0.5, 2.0)), sometimes( iaa.ElasticTransformation(alpha=(0.1, 2.0), sigma=0.25)), ], random_order=True) ], random_order=True)
def __init__(self): sometimes = lambda aug: iaa.Sometimes(0.5, aug) self.seq = iaa.Sequential([ iaa.OneOf([ iaa.Multiply((0.6, 1.0), per_channel=0.5), iaa.Multiply((1.0, 2.0), per_channel=0.5), ]), sometimes( iaa.OneOf([ iaa.Dropout((0.02, 0.03)), iaa.Salt((0.02, 0.03)) ]) ), sometimes(iaa.ChannelShuffle(1.0)), #sometimes(iaa.Invert(1.0, per_channel=True)), #sometimes(iaa.Invert(1.0)), #sometimes(iaa.CropAndPad( # percent=(-0.1, 0.1), pad_cval=(0,255) #)), iaa.Affine( #scale={"x": (0.8, 1.1), "y": (0.8, 1.1)}, # scale images to 80-120% of their size, individually per axis rotate=(-10, 10), # rotate by -45 to +45 degrees mode=ia.ALL # use any of scikit-image's warping modes (see 2nd image from the top for examples) ), sometimes(iaa.PerspectiveTransform(scale=(0.01, 0.04))), sometimes(iaa.AdditiveGaussianNoise((0.02, 0.1))), sometimes(iaa.AdditivePoissonNoise((0.02,0.1))), #sometimes(iaa.Pad( # percent=(0, 0.15), pad_mode=["edge"] #)) ])
def augment(image, bbox): x = random.randint(-60, 60) y = random.randint(-60, 60) aug = iaa.Sequential([iaa.AdditiveGaussianNoise(scale=random.uniform(.001, .01) * 255), # gaussian noise iaa.Multiply(random.uniform(0.5, 1.5)), # brightness iaa.Affine(translate_px={"x": x, "y": y}, # translation scale=random.uniform(0.5, 1.5), # zoom in and out rotate=random.uniform(-25, 25), # rotation shear=random.uniform(-5, 5), # shear transformation cval=(0, 255))]) # fill the empty space with color aug.add(iaa.Salt(.001)) bbs = ia.BoundingBoxesOnImage([ia.BoundingBox(x1=bbox[0], y1=bbox[1], x2=bbox[2], y2=bbox[3])], shape=image.shape) aug = aug.to_deterministic() image_aug = aug.augment_image(image) bbs_aug = aug.augment_bounding_boxes([bbs])[0] b = bbs_aug.bounding_boxes bbs_aug = [b[0].x1, b[0].y1, b[0].x2, b[0].y2] bbs_aug = np.asarray(bbs_aug) bbs_aug[0] = bbs_aug[0] if bbs_aug[0] > 0 else 0 bbs_aug[1] = bbs_aug[1] if bbs_aug[1] > 0 else 0 bbs_aug[2] = bbs_aug[2] if bbs_aug[2] < size else size bbs_aug[3] = bbs_aug[3] if bbs_aug[3] < size else size return image_aug, bbs_aug
def image_aug(image): """ @param image: @return: """ seq = iaa.SomeOf( (1, 3), [ iaa.Crop(px=(0, 16)), # 裁剪 iaa.Multiply((0.7, 1.3)), # 改变色调 iaa.Affine(scale=(0.5, 0.7)), # 放射变换 iaa.GaussianBlur(sigma=(0, 1.5)), # 高斯模糊 iaa.AddToHueAndSaturation(value=(25, -25)), iaa.ChannelShuffle(1), # RGB三通道随机交换 iaa.ElasticTransformation(alpha=0.1), iaa.Grayscale(alpha=(0.2, 0.5)), iaa.Pepper(p=0.03), iaa.AdditiveGaussianNoise(scale=(0.03 * 255, 0.05 * 255)), iaa.Dropout(p=(0.03, 0.05)), iaa.Salt(p=(0.03, 0.05)), iaa.AverageBlur(k=(1, 3)), iaa.Add((-10, 10)), iaa.CoarseSalt(size_percent=0.01) ]) seq_det = seq.to_deterministic() image_aug = seq_det.augment_images([image])[0] return image_aug
def generator(image_list): for name in image_list: fileName = name name = os.path.join(PATH, name) images = cv2.imread(name) sometimes = lambda aug: iaa.Sometimes(0.3, aug) seq = iaa.Sequential([ iaa.Flipud(p=0.5), iaa.Fliplr(p=0.5), sometimes(iaa.Pepper(p=0.10)), sometimes(iaa.Salt(p=0.03)), sometimes(iaa.AdditivePoissonNoise(lam=8.0)), sometimes(iaa.JpegCompression(compression=50)), sometimes(iaa.PiecewiseAffine(scale=0.015)), sometimes(iaa.MotionBlur(k=7, angle=0)), sometimes(iaa.MotionBlur(k=5, angle=144)) ], random_order=False) for i in range(10): images_aug = seq.augment_image(images) name = 'aug_' + fileName.split('.')[0] + "-" + str(i) + '.jpg' name = os.path.join(PATH, name) cv2.imwrite(name, images_aug) print(name + " is saved.")
def imgaugRGB(img): print(img.shape) seq = iaa.Sequential( [ # blur iaa.SomeOf((0, 2), [ iaa.GaussianBlur((0.0, 2.0)), iaa.AverageBlur(k=(3, 7)), iaa.MedianBlur(k=(3, 7)), iaa.BilateralBlur(d=(1, 7)), iaa.MotionBlur(k=(3, 7)) ]), #color iaa.SomeOf( (0, 2), [ #iaa.WithColorspace(), iaa.AddToHueAndSaturation((-20, 20)), #iaa.ChangeColorspace(to_colorspace[], alpha=0.5), iaa.Grayscale(alpha=(0.0, 0.2)) ]), #brightness iaa.OneOf([ iaa.Sequential([ iaa.Add((-10, 10), per_channel=0.5), iaa.Multiply((0.5, 1.5), per_channel=0.5) ]), iaa.Add((-10, 10), per_channel=0.5), iaa.Multiply((0.5, 1.5), per_channel=0.5), iaa.FrequencyNoiseAlpha(exponent=(-4, 0), first=iaa.Multiply( (0.5, 1.5), per_channel=0.5), second=iaa.ContrastNormalization( (0.5, 2.0), per_channel=0.5)) ]), #contrast iaa.SomeOf((0, 2), [ iaa.GammaContrast((0.5, 1.5), per_channel=0.5), iaa.SigmoidContrast( gain=(0, 10), cutoff=(0.25, 0.75), per_channel=0.5), iaa.LogContrast(gain=(0.75, 1), per_channel=0.5), iaa.LinearContrast(alpha=(0.7, 1.3), per_channel=0.5) ]), #arithmetic iaa.SomeOf((0, 3), [ iaa.AdditiveGaussianNoise(scale=(0, 0.05), per_channel=0.5), iaa.AdditiveLaplaceNoise(scale=(0, 0.05), per_channel=0.5), iaa.AdditivePoissonNoise(lam=(0, 8), per_channel=0.5), iaa.Dropout(p=(0, 0.05), per_channel=0.5), iaa.ImpulseNoise(p=(0, 0.05)), iaa.SaltAndPepper(p=(0, 0.05)), iaa.Salt(p=(0, 0.05)), iaa.Pepper(p=(0, 0.05)) ]), #iaa.Sometimes(p=0.5, iaa.JpegCompression((0, 30)), None), ], random_order=True) return seq.augment_image(img)
def main(): # datapath为存放训练图片的地方 datapath = '/home/zhex/data/OID_origin/train/Umbrella/' # original_file为需要被增强的 original_file = '/home/zhex/data/OID_origin/tools/new_txt/Umbrella.txt' # 需要被增强的训练真值txt # aug_file只记录了新增的增强后图片的box,要得到原始+增强的所有label:cat original_file augfile>finalfile(txt拼接) # aug_file输出是pdpd需要的格式,pytorch需要另行转换(可以拼接得到finalfile后直接将finalfile转换) aug_file = 'augfile_Umbrella.txt' dict_before = readlist(original_file) new_fp = open(aug_file, 'w') # augscene = {'Umbrellad':10,'hat':2} # 需要哪些场景,新增几倍数量的新数据 augscene = {'Umbrella': 5} for scene in augscene: # scene = Umbrella img_id = scene for i in range(augscene[scene]): for img_id in dict_before.keys(): img = Image.open(datapath + img_id) img = np.array(img) bbs = ia.BoundingBoxesOnImage([ ia.BoundingBox(x1=x, y1=y, x2=x + w, y2=y + h) for [x, y, w, h] in dict_before[img_id] ], shape=img.shape) # 设置数据增强方式 seq = iaa.SomeOf( (1, 3), [ iaa.Crop(px=(0, 16)), #裁剪 iaa.Multiply((0.7, 1.3)), #改变色调 iaa.Affine(scale=(0.5, 0.7)), #放射变换 iaa.GaussianBlur(sigma=(0, 1.5)), #高斯模糊 # iaa.AddToHueAndSaturation(value=(25,-25)), iaa.ChannelShuffle(1), # RGB三通道随机交换 iaa.ElasticTransformation(alpha=0.1), # iaa.Grayscale(alpha=(0.2, 0.5)), iaa.Pepper(p=0.03), iaa.AdditiveGaussianNoise(scale=(0.03 * 255, 0.05 * 255)), iaa.Dropout(p=(0.03, 0.05)), iaa.Salt(p=(0.03, 0.05)), iaa.AverageBlur(k=(1, 3)), iaa.Add((-10, 10)), iaa.CoarseSalt(size_percent=0.01) ]) seq_det = seq.to_deterministic( ) # 保持坐标和图像同步改变,每个batch都要调用一次,不然每次的增强都是一样的 image_aug = seq_det.augment_images([img])[0] bbs_aug = seq_det.augment_bounding_boxes([bbs])[0] pic_name = img_id.split('.')[0] # datapath = '/home/zhex/OID/train/Umbrella' if not os.path.exists(datapath + 'myaug/'): os.makedirs(datapath + 'myaug/') new_img_id = 'myaug/' + pic_name + '_{}'.format(i) + '.jpg' print('new_img_id = ', new_img_id) Image.fromarray(image_aug).save(datapath + new_img_id) new_fp = writelist(new_fp, new_img_id, bbs_aug.bounding_boxes)
def chapter_augmenters_salt(): fn_start = "arithmetic/salt" aug = iaa.Salt(0.1) run_and_save_augseq(fn_start + ".jpg", aug, [ia.quokka(size=(128, 128)) for _ in range(8)], cols=4, rows=2, quality=95)
def transform_op(transform): if transform == 'flip_h': seq = iaa.Sequential([ iaa.Fliplr(1), ]) elif transform == 'invert': seq = iaa.Sequential([ iaa.Invert(1, True), ]) elif transform == 'brigth': seq = iaa.Sequential([ iaa.Multiply(1.5, True), ]) elif transform == 'dark': seq = iaa.Sequential([ iaa.Multiply(0.5, True), ]) elif transform == 'blur': seq = iaa.Sequential([ iaa.GaussianBlur(0.5), ]) elif transform == 'sharp': seq = iaa.Sequential([ iaa.Sharpen(1, 1.25), ]) elif transform == 'dark_sharp': seq = iaa.Sequential([ iaa.Sharpen(1, 0.25), ]) elif transform == 'gauss_noise': seq = iaa.Sequential([ iaa.AdditiveGaussianNoise(0.03, 10, True), ]) elif transform == 'dropout': seq = iaa.Sequential([ iaa.Dropout(0.08, True), ]) elif transform == 'salt': seq = iaa.Sequential([ iaa.Salt(0.08, True), ]) elif transform == 'salt_pepper': seq = iaa.Sequential([ iaa.SaltAndPepper(0.08, True), ]) elif transform == 'contrast': seq = iaa.Sequential([ iaa.ContrastNormalization(1.5, True), ]) else: # So far it's only implemented horizontal flips print("Sorry, only those operations listed in help are implemented. \ Check data_augmentation.py -help for further instructions") return None return seq
def __getitem__(self, idx): img_pil = Image.open( os.path.join(self.img_dir, self.img_name[idx] + '.jpg')).convert('RGB') # read targets, coordinates: x1, y1, x2, y2 targets = pd.read_csv( os.path.join(self.label_dir, self.img_name[idx] + '.txt'), sep=" ", header=None, names=['class', 'left', 'top', 'right', 'bottom']) targets_boxes = targets[['left', 'top', 'right', 'bottom']] # number of object on image num_objs = len(targets_boxes) # get bounding box for each object on image list_boxes = [] for i in range(num_objs): list_boxes.append(targets_boxes.iloc[i].tolist()) # convert boxes to tensor boxes = torch.as_tensor(list_boxes, dtype=torch.float32) # there is only one class labels = torch.ones((num_objs, ), dtype=torch.int64) # img transform img = self.img_transform(img_pil) target = {} target['boxes'] = boxes target['labels'] = labels if self.train: # convert image to numpy array img = np.array(img_pil) # define augmentation # applies the given augmenter in 50% of all cases sometimes = lambda aug: iaa.Sometimes(0.3, aug) sequence = iaa.Sequential([ iaa.Resize(size=224), sometimes(iaa.Salt(p=0.10)), sometimes(iaa.Multiply((0.25, 0.50))) ]) # boxes transform bbs = self._get_list_bbs(list_boxes) boxes = ia.BoundingBoxesOnImage(bounding_boxes=bbs, shape=img.shape) boxes = sequence.augment_bounding_boxes(boxes) boxes = boxes.to_xyxy_array(dtype=np.float32) boxes = torch.from_numpy(boxes) target['boxes'] = boxes # image transform transform = self.img_transform img = sequence.augment_image(img) img = transform(img.copy()) return img, target
def generate_background_noise(image_shape): random_state = int(np.random.rand() * 100000) background_noise = np.zeros(image_shape, dtype=np.uint8) background_noise[:] = [0, 0, 0, 255] noise_augmentation = iaa.Sequential([ iaa.Salt(p=0.041, random_state=random_state), iaa.AdditiveGaussianNoise(scale=0.1 * 255), iaa.MotionBlur(angle=90, k=8), iaa.GaussianBlur(sigma=0.6) ]) return noise_augmentation(image=background_noise)
def aug_before_prepare(self, img, bboxes, masks, polys): aug = iaa.SomeOf((1, 3), [ iaa.Multiply(((0.7, 1.3))), iaa.GaussianBlur(sigma=(0, 1.5)), iaa.ChannelShuffle(1), iaa.Pepper(p=0.03), iaa.AdditiveGaussianNoise(scale=(0.03 * 255, 0.05 * 255)), iaa.Dropout(p=(0.03, 0.05)), iaa.Salt(p=(0.03, 0.05)), iaa.Add((-10, 10)), iaa.AverageBlur(k=(1, 3)) ]) if (random.random() > 0.5): img, bboxes, masks = self.random_rotate(img, np.array(polys), masks, 90) img = aug.augment_image(img) return img, bboxes, masks
def salt(image, prob, keys): """ Adding salt noise """ r = random.uniform(1, 5) * 0.05 aug = iaa.Sequential([ iaa.Dropout(p=(0, r)), iaa.CoarseDropout(p=0.001, size_percent=0.01), iaa.Salt(0.001), iaa.AdditiveGaussianNoise(scale=0.1 * 255) ]) aug.add(iaa.Multiply(random.uniform(0.25, 1.5))) x = random.randrange(-10, 10) * .01 y = random.randrange(-10, 10) * .01 aug.add( iaa.Affine(scale=random.uniform(.7, 1.1), translate_percent={ "x": x, "y": y }, cval=(0, 255))) seq_det = aug.to_deterministic() image_aug = seq_det.augment_images([image])[0] keys = ia.KeypointsOnImage([ ia.Keypoint(x=keys[0], y=keys[1]), ia.Keypoint(x=keys[2], y=keys[3]), ia.Keypoint(x=keys[4], y=keys[5]), ia.Keypoint(x=keys[6], y=keys[7]), ia.Keypoint(x=keys[8], y=keys[9]) ], shape=image.shape) keys_aug = seq_det.augment_keypoints([keys])[0] k = keys_aug.keypoints output = [ k[0].x, k[0].y, k[1].x, k[1].y, k[2].x, k[2].y, k[3].x, k[3].y, k[4].x, k[4].y ] index = 0 for i in range(0, len(prob)): output[index] = output[index] * prob[i] output[index + 1] = output[index + 1] * prob[i] index = index + 2 output = np.array(output) return image_aug, output
def do_all_aug(image): do_aug(image, iaa.Noop(name="origin")) do_aug(image, iaa.Crop((0, 10))) # 切边 do_aug(image, iaa.GaussianBlur((0, 3))) do_aug(image, iaa.AverageBlur(1, 7)) do_aug(image, iaa.MedianBlur(1, 7)) do_aug(image, iaa.Sharpen()) do_aug(image, iaa.BilateralBlur()) # 既噪音又模糊,叫双边 do_aug(image, iaa.MotionBlur()) do_aug(image, iaa.MeanShiftBlur()) do_aug(image, iaa.GammaContrast()) do_aug(image, iaa.SigmoidContrast()) do_aug(image, iaa.Affine(shear={ 'x': (-10, 10), 'y': (-10, 10) }, mode="edge")) # shear:x轴往左右偏离的像素书,(a,b)是a,b间随机值,[a,b]是二选一 do_aug(image, iaa.Affine(shear={ 'x': (-10, 10), 'y': (-10, 10) }, mode="edge")) # shear:x轴往左右偏离的像素书,(a,b)是a,b间随机值,[a,b]是二选一 do_aug(image, iaa.Rotate(rotate=(-10, 10), mode="edge")) do_aug(image, iaa.PiecewiseAffine()) # 局部点变形 do_aug(image, iaa.Fog()) do_aug(image, iaa.Clouds()) do_aug(image, iaa.Snowflakes(flake_size=(0.1, 0.2), density=(0.005, 0.025))) do_aug( image, iaa.Rain( nb_iterations=1, drop_size=(0.05, 0.1), speed=(0.04, 0.08), )) do_aug( image, iaa.ElasticTransformation(alpha=(0.0, 20.0), sigma=(3.0, 5.0), mode="nearest")) do_aug(image, iaa.AdditiveGaussianNoise(scale=(0, 10))) do_aug(image, iaa.AdditiveLaplaceNoise(scale=(0, 10))) do_aug(image, iaa.AdditivePoissonNoise(lam=(0, 10))) do_aug(image, iaa.Salt((0, 0.02))) do_aug(image, iaa.Pepper((0, 0.02)))
def __init__(self): sometimes = lambda aug: iaa.Sometimes(0.5, aug) self.seq = iaa.Sequential([ sometimes(iaa.Crop(px=(0, 0, 8, 0), keep_size=True)), sometimes(iaa.Pad(px=(0, 0, 0, 5), keep_size=False)), iaa.Multiply((0.8, 1.2), per_channel=0.5), sometimes(iaa.PerspectiveTransform(scale=(0.01, 0.05))), sometimes( iaa.OneOf([ iaa.CoarseDropout((0.01, 0.03), size_percent=(0.1, 0.3)), iaa.CoarseDropout((0.01, 0.03), size_percent=(0.1, 0.3), per_channel=1.0), iaa.Dropout((0.03,0.05)), iaa.Salt((0.03,0.05)) ]) ), iaa.Multiply((0.8, 1.2), per_channel=0.5), sometimes(iaa.FrequencyNoiseAlpha( exponent=(-4, 0), first=iaa.Multiply((0.8, 1.2), per_channel=0.5), second=iaa.ContrastNormalization((0.8, 1.5)) ) ), sometimes( iaa.OneOf([ iaa.MotionBlur(k=(3,4),angle=(0, 360)), iaa.GaussianBlur((0, 1.2)), iaa.AverageBlur(k=(2, 3)), iaa.MedianBlur(k=(3, 5)) ]) ), sometimes( iaa.CropAndPad( percent=(-0.05, 0.1), pad_mode='constant', pad_cval=(0, 255) ), ), sometimes(iaa.ElasticTransformation(alpha=(1.0, 2.0), sigma=(2.0, 3.0))), # move pixels locally around (with random strengths) sometimes(iaa.PiecewiseAffine(scale=(0.01, 0.02), mode='constant')), # sometimes move parts of the image around sometimes(iaa.AdditiveGaussianNoise((0.02, 0.1))), sometimes(iaa.AdditivePoissonNoise((0.02,0.05))), iaa.Invert(p=0.5) ])
def __init__(self): sometimes = lambda aug: iaa.Sometimes(0.5, aug) self.seq = iaa.Sequential([ iaa.Multiply((0.8, 1.2), per_channel=0.5), sometimes( iaa.OneOf([ iaa.CoarseDropout((0.01, 0.03), size_percent=(0.1, 0.3)), iaa.CoarseDropout((0.01, 0.03), size_percent=(0.1, 0.3), per_channel=1.0), iaa.Dropout((0.03,0.05)), iaa.Salt((0.03,0.05)) ]) ), sometimes(iaa.FrequencyNoiseAlpha( exponent=(-4, 0), first=iaa.Multiply((0.8, 1.2), per_channel=0.5), second=iaa.ContrastNormalization((0.8, 1.5)) ) ), sometimes( iaa.OneOf([ iaa.MotionBlur(k=(3,4),angle=(0, 360)), iaa.GaussianBlur((0, 1.2)), iaa.AverageBlur(k=(2, 3)), iaa.MedianBlur(k=(3, 5)) ]) ), sometimes( iaa.CropAndPad( percent=(-0.02, 0.02), pad_mode='constant', pad_cval=(0, 255) ), ), sometimes(iaa.AdditiveGaussianNoise((0.02, 0.1))), sometimes(iaa.AdditivePoissonNoise((0.02,0.05))), iaa.Invert(p=0.5) ])
def salt(image, keys): """ Adding noise """ r = random.randint(1, 5) * 0.1 aug = iaa.Sequential([ iaa.Dropout(p=(0, r)), iaa.CoarseDropout(p=0.02, size_percent=0.5), iaa.Salt(0.05) ]) aug.add(iaa.Multiply(random.uniform(0.35, 1.5))) aug.add(iaa.Affine(rotate=random.randint(-180, 180))) aug.add(iaa.Affine(scale=random.uniform(.2, 1.2), cval=(0, 255))) seq_det = aug.to_deterministic() keys = ia.KeypointsOnImage( [ia.Keypoint(x=keys[0], y=keys[1]), ia.Keypoint(x=keys[2], y=keys[3])], shape=image.shape) image_aug = seq_det.augment_images([image])[0] keys_aug = seq_det.augment_keypoints([keys])[0] k = keys_aug.keypoints keys_aug = [k[0].x, k[0].y, k[1].x, k[1].y] keys_aug = np.asarray(keys_aug) return image_aug, keys_aug
# for the rest from the same poisson distribution: "Additive_Poisson_Noise": lambda lo, hi, percent: iaa.AdditivePoissonNoise(lam=(lo, hi), per_channel=percent), # Adds salt and pepper noise to an image, i.e. some white-ish and black-ish pixels. # Replaces percent of all pixels with salt and pepper noise "Salt_And_Pepper": lambda percent: iaa.SaltAndPepper(percent), # Adds coarse salt and pepper noise to image, i.e. rectangles that contain noisy white-ish and black-ish pixels # Replaces percent of all pixels with salt/pepper in an image that has lo to hi percent of the input image size, # then upscales the results to the input image size, leading to large rectangular areas being replaced. "Coarse_Salt_And_Pepper": lambda percent, lo, hi: iaa.CoarseSaltAndPepper(percent, size_percent=(lo, hi)), # Adds salt noise to an image, i.e white-ish pixels # Replaces percent of all pixels with salt noise "Salt": lambda percent: iaa.Salt(percent), # Adds coarse salt noise to image, i.e. rectangles that contain noisy white-ish pixels # Replaces percent of all pixels with salt in an image that has lo to hi percent of the input image size, # then upscales the results to the input image size, leading to large rectangular areas being replaced. "Coarse_Salt": lambda percent, lo, hi: iaa.CoarseSalt(percent, size_percent=(lo, hi)), # Adds Pepper noise to an image, i.e Black-ish pixels # Replaces percent of all pixels with Pepper noise "Pepper": lambda percent: iaa.Pepper(percent), # Adds coarse pepper noise to image, i.e. rectangles that contain noisy black-ish pixels # Replaces percent of all pixels with salt in an image that has lo to hi percent of the input image size, # then upscales the results to the input image size, leading to large rectangular areas being replaced. "Coarse_Pepper": lambda percent, lo, hi: iaa.CoarsePepper(percent, size_percent=(lo, hi)),
def salt(img): seq = iaa.Sequential( [iaa.Salt(per_channel=True, p=random.uniform(0, 0.5))]) return seq.augment_image(img)
#!/usr/bin/env python #-*- coding:utf-8 -*- #author: wu.zheng midday.me from imgaug import augmenters as iaa import cv2 import imgaug as ia from imgaug.augmentables.segmaps import SegmentationMapOnImage from imgaug.augmentables.bbs import BoundingBox, BoundingBoxesOnImage import random seq = iaa.SomeOf((1, 4), [ iaa.Salt(p=(0.2, 0.4)), iaa.GaussianBlur(sigma=(0, 2.0)), iaa.CoarseDropout(p=(0.02, 0.1), size_percent=(0.2, 0.6)), iaa.JpegCompression(compression=(30, 50)), ]) def augment_with_segmap(image, segmap, num_classes): if random.random() < 0.3: return image, segmap segmap = SegmentationMapOnImage(segmap, shape=image.shape, nb_classes=num_classes) image_aug, segmap_aug = seq(image=image, segmentation_maps=segmap) return image_aug, None
def main(): parser = argparse.ArgumentParser(description="Check augmenters visually.") parser.add_argument( "--only", default=None, help= "If this is set, then only the results of an augmenter with this name will be shown. " "Optionally, comma-separated list.", required=False) args = parser.parse_args() images = [ ia.quokka_square(size=(128, 128)), ia.imresize_single_image(data.astronaut(), (128, 128)) ] keypoints = [ ia.KeypointsOnImage([ ia.Keypoint(x=50, y=40), ia.Keypoint(x=70, y=38), ia.Keypoint(x=62, y=52) ], shape=images[0].shape), ia.KeypointsOnImage([ ia.Keypoint(x=55, y=32), ia.Keypoint(x=42, y=95), ia.Keypoint(x=75, y=89) ], shape=images[1].shape) ] bounding_boxes = [ ia.BoundingBoxesOnImage([ ia.BoundingBox(x1=10, y1=10, x2=20, y2=20), ia.BoundingBox(x1=40, y1=50, x2=70, y2=60) ], shape=images[0].shape), ia.BoundingBoxesOnImage([ ia.BoundingBox(x1=10, y1=10, x2=20, y2=20), ia.BoundingBox(x1=40, y1=50, x2=70, y2=60) ], shape=images[1].shape) ] augmenters = [ iaa.Sequential([ iaa.CoarseDropout(p=0.5, size_percent=0.05), iaa.AdditiveGaussianNoise(scale=0.1 * 255), iaa.Crop(percent=0.1) ], name="Sequential"), iaa.SomeOf(2, children=[ iaa.CoarseDropout(p=0.5, size_percent=0.05), iaa.AdditiveGaussianNoise(scale=0.1 * 255), iaa.Crop(percent=0.1) ], name="SomeOf"), iaa.OneOf(children=[ iaa.CoarseDropout(p=0.5, size_percent=0.05), iaa.AdditiveGaussianNoise(scale=0.1 * 255), iaa.Crop(percent=0.1) ], name="OneOf"), iaa.Sometimes(0.5, iaa.AdditiveGaussianNoise(scale=0.1 * 255), name="Sometimes"), iaa.WithColorspace("HSV", children=[iaa.Add(20)], name="WithColorspace"), iaa.WithChannels([0], children=[iaa.Add(20)], name="WithChannels"), iaa.AddToHueAndSaturation((-20, 20), per_channel=True, name="AddToHueAndSaturation"), iaa.Noop(name="Noop"), iaa.Resize({ "width": 64, "height": 64 }, name="Resize"), iaa.CropAndPad(px=(-8, 8), name="CropAndPad-px"), iaa.Pad(px=(0, 8), name="Pad-px"), iaa.Crop(px=(0, 8), name="Crop-px"), iaa.Crop(percent=(0, 0.1), name="Crop-percent"), iaa.Fliplr(0.5, name="Fliplr"), iaa.Flipud(0.5, name="Flipud"), iaa.Superpixels(p_replace=0.75, n_segments=50, name="Superpixels"), iaa.Grayscale(0.5, name="Grayscale0.5"), iaa.Grayscale(1.0, name="Grayscale1.0"), iaa.GaussianBlur((0, 3.0), name="GaussianBlur"), iaa.AverageBlur(k=(3, 11), name="AverageBlur"), iaa.MedianBlur(k=(3, 11), name="MedianBlur"), iaa.BilateralBlur(d=10, name="BilateralBlur"), iaa.Sharpen(alpha=(0.1, 1.0), lightness=(0, 2.0), name="Sharpen"), iaa.Emboss(alpha=(0.1, 1.0), strength=(0, 2.0), name="Emboss"), iaa.EdgeDetect(alpha=(0.1, 1.0), name="EdgeDetect"), iaa.DirectedEdgeDetect(alpha=(0.1, 1.0), direction=(0, 1.0), name="DirectedEdgeDetect"), iaa.Add((-50, 50), name="Add"), iaa.Add((-50, 50), per_channel=True, name="AddPerChannel"), iaa.AddElementwise((-50, 50), name="AddElementwise"), iaa.AdditiveGaussianNoise(loc=0, scale=(0.0, 0.1 * 255), name="AdditiveGaussianNoise"), iaa.Multiply((0.5, 1.5), name="Multiply"), iaa.Multiply((0.5, 1.5), per_channel=True, name="MultiplyPerChannel"), iaa.MultiplyElementwise((0.5, 1.5), name="MultiplyElementwise"), iaa.Dropout((0.0, 0.1), name="Dropout"), iaa.CoarseDropout(p=0.05, size_percent=(0.05, 0.5), name="CoarseDropout"), iaa.Invert(p=0.5, name="Invert"), iaa.Invert(p=0.5, per_channel=True, name="InvertPerChannel"), iaa.ContrastNormalization(alpha=(0.5, 2.0), name="ContrastNormalization"), iaa.SaltAndPepper(p=0.05, name="SaltAndPepper"), iaa.Salt(p=0.05, name="Salt"), iaa.Pepper(p=0.05, name="Pepper"), iaa.CoarseSaltAndPepper(p=0.05, size_percent=(0.01, 0.1), name="CoarseSaltAndPepper"), iaa.CoarseSalt(p=0.05, size_percent=(0.01, 0.1), name="CoarseSalt"), iaa.CoarsePepper(p=0.05, size_percent=(0.01, 0.1), name="CoarsePepper"), iaa.Affine(scale={ "x": (0.8, 1.2), "y": (0.8, 1.2) }, translate_px={ "x": (-16, 16), "y": (-16, 16) }, rotate=(-45, 45), shear=(-16, 16), order=ia.ALL, cval=(0, 255), mode=ia.ALL, name="Affine"), iaa.PiecewiseAffine(scale=0.03, nb_rows=(2, 6), nb_cols=(2, 6), name="PiecewiseAffine"), iaa.PerspectiveTransform(scale=0.1, name="PerspectiveTransform"), iaa.ElasticTransformation(alpha=(0.5, 8.0), sigma=1.0, name="ElasticTransformation"), iaa.Alpha(factor=(0.0, 1.0), first=iaa.Add(100), second=iaa.Dropout(0.5), per_channel=False, name="Alpha"), iaa.Alpha(factor=(0.0, 1.0), first=iaa.Add(100), second=iaa.Dropout(0.5), per_channel=True, name="AlphaPerChannel"), iaa.Alpha(factor=(0.0, 1.0), first=iaa.Affine(rotate=(-45, 45)), per_channel=True, name="AlphaAffine"), iaa.AlphaElementwise(factor=(0.0, 1.0), first=iaa.Add(50), second=iaa.ContrastNormalization(2.0), per_channel=False, name="AlphaElementwise"), iaa.AlphaElementwise(factor=(0.0, 1.0), first=iaa.Add(50), second=iaa.ContrastNormalization(2.0), per_channel=True, name="AlphaElementwisePerChannel"), iaa.AlphaElementwise(factor=(0.0, 1.0), first=iaa.Affine(rotate=(-45, 45)), per_channel=True, name="AlphaElementwiseAffine"), iaa.SimplexNoiseAlpha(first=iaa.EdgeDetect(1.0), per_channel=False, name="SimplexNoiseAlpha"), iaa.FrequencyNoiseAlpha(first=iaa.EdgeDetect(1.0), per_channel=False, name="FrequencyNoiseAlpha") ] augmenters.append( iaa.Sequential([iaa.Sometimes(0.2, aug.copy()) for aug in augmenters], name="Sequential")) augmenters.append( iaa.Sometimes(0.5, [aug.copy() for aug in augmenters], name="Sometimes")) for augmenter in augmenters: if args.only is None or augmenter.name in [ v.strip() for v in args.only.split(",") ]: print("Augmenter: %s" % (augmenter.name, )) grid = [] for image, kps, bbs in zip(images, keypoints, bounding_boxes): aug_det = augmenter.to_deterministic() imgs_aug = aug_det.augment_images( np.tile(image[np.newaxis, ...], (16, 1, 1, 1))) kps_aug = aug_det.augment_keypoints([kps] * 16) bbs_aug = aug_det.augment_bounding_boxes([bbs] * 16) imgs_aug_drawn = [ kps_aug_one.draw_on_image(img_aug) for img_aug, kps_aug_one in zip(imgs_aug, kps_aug) ] imgs_aug_drawn = [ bbs_aug_one.draw_on_image(img_aug) for img_aug, bbs_aug_one in zip(imgs_aug_drawn, bbs_aug) ] grid.append(np.hstack(imgs_aug_drawn)) ia.imshow(np.vstack(grid))
tip = amp._get_tip_queue() augmenters = [ # blur images with a sigma between 0 and 3.0 iaa.Noop(), iaa.GaussianBlur(sigma=(0.5, 2.0)), iaa.Add((-50.0, 50.0), per_channel=False), iaa.AdditiveGaussianNoise(loc=0, scale=(0.07 * 255, 0.07 * 255), per_channel=False), iaa.Dropout(p=0.07, per_channel=False), iaa.CoarseDropout(p=(0.05, 0.15), size_percent=(0.1, 0.9), per_channel=False), iaa.SaltAndPepper(p=(0.05, 0.15), per_channel=False), iaa.Salt(p=(0.05, 0.15), per_channel=False), iaa.Pepper(p=(0.05, 0.15), per_channel=False), iaa.ContrastNormalization(alpha=(iap.Uniform(0.02, 0.03), iap.Uniform(1.7, 2.1))), iaa.ElasticTransformation(alpha=(0.5, 2.0)), ] seq = iaa.Sequential(iaa.OneOf(augmenters), ) def get_data_from_tip(tip, batch_size): features = [] labels = [] descriptions = [] for i in range(batch_size): data = tip.get() d, f, l = data
def main(): # datapath为存放训练图片的地方 datapath = '/home/zhex/data/yuncong/' # original_file为需要被增强的 original_file = '/home/zhex/data/yuncong/Mall_train.txt' # 需要被增强的训练真值txt # aug_file只记录了新增的增强后图片的box,要得到原始+增强的所有label:cat original_file augfile>finalfile(txt拼接) # aug_file输出是pdpd需要的格式,pytorch需要另行转换(可以拼接得到finalfile后直接将finalfile转换) aug_file = 'augfile_Mall.txt' dict_before = readlist(original_file) new_fp = open(aug_file, 'w') # augscene = {'Mall': 3, 'Part_B': 10, 'Part_A': 13} # 需要哪些场景,新增几倍数量的新数据 augscene = {'Mall': 3} for scene in augscene: for i in range(augscene[scene]): for img_id in dict_before.keys(): if scene in img_id: print(img_id) img = Image.open( datapath + img_id) # img.convert('RGB')->img.save('filename.jpg') img = np.array(img) bbs = ia.BoundingBoxesOnImage([ ia.BoundingBox(x1=x, y1=y, x2=x + w, y2=y + h) for [x, y, w, h] in dict_before[img_id] ], shape=img.shape) # 设置数据增强方式 # import imgaug.augmenters as iaa # List augmenter that applies only some of its children to images ''' iaa.SomeOf(n=None, children=None, random_order=False, name=None, deterministic=False, random_state=None) n: 从总的Augmenters中选择多少个来处理图片,类型可以是int,tuple,list,或者随机值 random_order: 是否每次顺序一样,默认值False(即每次顺序一样) ''' seq = iaa.SomeOf( (1, 3), [ #每次使用1~3个Augmenter来处理图片,每个batch的顺序一样 iaa.Crop(px=(0, 16)), #裁剪 iaa.Multiply((0.7, 1.3)), #改变色调 iaa.Affine(scale=(0.5, 0.7)), #仿射变换 iaa.GaussianBlur(sigma=(0, 1.5)), #高斯模糊 iaa.AddToHueAndSaturation(value=(25, -25)), iaa.ChannelShuffle(1), # RGB三通道随机交换 iaa.ElasticTransformation(alpha=0.1), iaa.Grayscale(alpha=(0.2, 0.5)), iaa.Pepper(p=0.03), iaa.AdditiveGaussianNoise(scale=(0.03 * 255, 0.05 * 255)), iaa.Dropout(p=(0.03, 0.05)), iaa.Salt(p=(0.03, 0.05)), iaa.AverageBlur(k=(1, 3)), iaa.Add((-10, 10)), iaa.CoarseSalt(size_percent=0.01) ], random_order=False) seq_det = seq.to_deterministic( ) # 保持坐标和图像同步改变,每个batch都要调用一次,不然每次的增强都是一样的 image_aug = seq_det.augment_images([img])[0] bbs_aug = seq_det.augment_bounding_boxes([bbs])[0] pic_name = img_id.split('/')[-1].split('.')[0] pic_dir = img_id.split(pic_name)[0] if not os.path.exists(datapath + 'myaug/' + pic_dir): os.makedirs(datapath + 'myaug/' + pic_dir) new_img_id = 'myaug/' + pic_dir + pic_name + '_{}'.format( i) + '.jpg' Image.fromarray(image_aug).save(datapath + new_img_id) new_fp = writelist(new_fp, new_img_id, bbs_aug.bounding_boxes)
def augment(train_x, train_y_prob, train_y_keys): aug = iaa.Sequential() x = random.uniform(-10, 10) * .01 y = random.uniform(-10, 10) * .01 aug.add(iaa.Affine(translate_percent={"x": x, "y": y}, # (-10, 10) -> (-1, 3) scale=random.uniform(.7, 1.1), # (.7, 1.1) -> (.99, 1.01) rotate=random.uniform(-10, 10), # (-10, 10) -> (-5, 5) shear=random.uniform(-15, 15), # (-15, 15) -> (.5, .5) cval=(0, 255))) # (0 ~ 255) """ random brightness """ brightness = random.uniform(.5, 1.5) # (.5, 1.5) -> (.995, 1.005) aug.add(iaa.Multiply(brightness)) """ random pixel dropout """ aug.add(iaa.CoarseDropout(p=.001, size_percent=0.005)) aug.add(iaa.Dropout(p=(0, random.uniform(1, 5) * 0.005))) """ salt noise """ aug.add(iaa.Salt(.001)) """ additive gaussian noise """ aug.add(iaa.AdditiveGaussianNoise(scale=random.uniform(.01, .1) * 255)) """ crop """ pixel = 5 aug.add(iaa.Crop(px=((0, random.randint(0, pixel)), (0, random.randint(0, pixel)), (0, random.randint(0, pixel)), (0, random.randint(0, pixel))))) seq_det = aug.to_deterministic() image_aug = [] keys_aug = [] for i in range(0, train_x.shape[0]): image = train_x[i, :, :, :] prob = train_y_prob[i, :] keys = train_y_keys[i, :] image_aug.append(seq_det.augment_images([image])[0]) koi = ia.KeypointsOnImage([ia.Keypoint(x=keys[0], y=keys[1]), ia.Keypoint(x=keys[2], y=keys[3]), ia.Keypoint(x=keys[4], y=keys[5]), ia.Keypoint(x=keys[6], y=keys[7]), ia.Keypoint(x=keys[8], y=keys[9])], shape=image.shape) k = seq_det.augment_keypoints([koi])[0] k = k.keypoints keys = [k[0].x, k[0].y, k[1].x, k[1].y, k[2].x, k[2].y, k[3].x, k[3].y, k[4].x, k[4].y] index = 0 prob = prob.T[0][:-1] # last index is a direction parameter for j in range(0, len(prob)): keys[index] = keys[index] * prob[j] keys[index + 1] = keys[index + 1] * prob[j] index = index + 2 keys_aug.append(keys) image_aug = np.asarray(image_aug) keys_aug = np.asarray(keys_aug) keys_aug = np.expand_dims(keys_aug, axis=-1) return image_aug, keys_aug
def create_augmenters(height, width, height_augmentable, width_augmentable, only_augmenters): def lambda_func_images(images, random_state, parents, hooks): return images def lambda_func_heatmaps(heatmaps, random_state, parents, hooks): return heatmaps def lambda_func_keypoints(keypoints, random_state, parents, hooks): return keypoints def assertlambda_func_images(images, random_state, parents, hooks): return True def assertlambda_func_heatmaps(heatmaps, random_state, parents, hooks): return True def assertlambda_func_keypoints(keypoints, random_state, parents, hooks): return True augmenters_meta = [ iaa.Sequential([iaa.Noop(), iaa.Noop()], random_order=False, name="Sequential_2xNoop"), iaa.Sequential([iaa.Noop(), iaa.Noop()], random_order=True, name="Sequential_2xNoop_random_order"), iaa.SomeOf((1, 3), [iaa.Noop(), iaa.Noop(), iaa.Noop()], random_order=False, name="SomeOf_3xNoop"), iaa.SomeOf((1, 3), [iaa.Noop(), iaa.Noop(), iaa.Noop()], random_order=True, name="SomeOf_3xNoop_random_order"), iaa.OneOf([iaa.Noop(), iaa.Noop(), iaa.Noop()], name="OneOf_3xNoop"), iaa.Sometimes(0.5, iaa.Noop(), name="Sometimes_Noop"), iaa.WithChannels([1, 2], iaa.Noop(), name="WithChannels_1_and_2_Noop"), iaa.Noop(name="Noop"), iaa.Lambda(func_images=lambda_func_images, func_heatmaps=lambda_func_heatmaps, func_keypoints=lambda_func_keypoints, name="Lambda"), iaa.AssertLambda(func_images=assertlambda_func_images, func_heatmaps=assertlambda_func_heatmaps, func_keypoints=assertlambda_func_keypoints, name="AssertLambda"), iaa.AssertShape((None, height_augmentable, width_augmentable, None), name="AssertShape"), iaa.ChannelShuffle(0.5, name="ChannelShuffle") ] augmenters_arithmetic = [ iaa.Add((-10, 10), name="Add"), iaa.AddElementwise((-10, 10), name="AddElementwise"), #iaa.AddElementwise((-500, 500), name="AddElementwise"), iaa.AdditiveGaussianNoise(scale=(5, 10), name="AdditiveGaussianNoise"), iaa.AdditiveLaplaceNoise(scale=(5, 10), name="AdditiveLaplaceNoise"), iaa.AdditivePoissonNoise(lam=(1, 5), name="AdditivePoissonNoise"), iaa.Multiply((0.5, 1.5), name="Multiply"), iaa.MultiplyElementwise((0.5, 1.5), name="MultiplyElementwise"), iaa.Dropout((0.01, 0.05), name="Dropout"), iaa.CoarseDropout((0.01, 0.05), size_percent=(0.01, 0.1), name="CoarseDropout"), iaa.ReplaceElementwise((0.01, 0.05), (0, 255), name="ReplaceElementwise"), #iaa.ReplaceElementwise((0.95, 0.99), (0, 255), name="ReplaceElementwise"), iaa.SaltAndPepper((0.01, 0.05), name="SaltAndPepper"), iaa.ImpulseNoise((0.01, 0.05), name="ImpulseNoise"), iaa.CoarseSaltAndPepper((0.01, 0.05), size_percent=(0.01, 0.1), name="CoarseSaltAndPepper"), iaa.Salt((0.01, 0.05), name="Salt"), iaa.CoarseSalt((0.01, 0.05), size_percent=(0.01, 0.1), name="CoarseSalt"), iaa.Pepper((0.01, 0.05), name="Pepper"), iaa.CoarsePepper((0.01, 0.05), size_percent=(0.01, 0.1), name="CoarsePepper"), iaa.Invert(0.1, name="Invert"), # ContrastNormalization iaa.JpegCompression((50, 99), name="JpegCompression") ] augmenters_blend = [ iaa.Alpha((0.01, 0.99), iaa.Noop(), name="Alpha"), iaa.AlphaElementwise((0.01, 0.99), iaa.Noop(), name="AlphaElementwise"), iaa.SimplexNoiseAlpha(iaa.Noop(), name="SimplexNoiseAlpha"), iaa.FrequencyNoiseAlpha((-2.0, 2.0), iaa.Noop(), name="FrequencyNoiseAlpha") ] augmenters_blur = [ iaa.GaussianBlur(sigma=(1.0, 5.0), name="GaussianBlur"), iaa.AverageBlur(k=(3, 11), name="AverageBlur"), iaa.MedianBlur(k=(3, 11), name="MedianBlur"), iaa.BilateralBlur(d=(3, 11), name="BilateralBlur"), iaa.MotionBlur(k=(3, 11), name="MotionBlur") ] augmenters_color = [ # InColorspace (deprecated) iaa.WithColorspace(to_colorspace="HSV", children=iaa.Noop(), name="WithColorspace"), iaa.WithHueAndSaturation(children=iaa.Noop(), name="WithHueAndSaturation"), iaa.MultiplyHueAndSaturation((0.8, 1.2), name="MultiplyHueAndSaturation"), iaa.MultiplyHue((-1.0, 1.0), name="MultiplyHue"), iaa.MultiplySaturation((0.8, 1.2), name="MultiplySaturation"), iaa.AddToHueAndSaturation((-10, 10), name="AddToHueAndSaturation"), iaa.AddToHue((-10, 10), name="AddToHue"), iaa.AddToSaturation((-10, 10), name="AddToSaturation"), iaa.ChangeColorspace(to_colorspace="HSV", name="ChangeColorspace"), iaa.Grayscale((0.01, 0.99), name="Grayscale"), iaa.KMeansColorQuantization((2, 16), name="KMeansColorQuantization"), iaa.UniformColorQuantization((2, 16), name="UniformColorQuantization") ] augmenters_contrast = [ iaa.GammaContrast(gamma=(0.5, 2.0), name="GammaContrast"), iaa.SigmoidContrast(gain=(5, 20), cutoff=(0.25, 0.75), name="SigmoidContrast"), iaa.LogContrast(gain=(0.7, 1.0), name="LogContrast"), iaa.LinearContrast((0.5, 1.5), name="LinearContrast"), iaa.AllChannelsCLAHE(clip_limit=(2, 10), tile_grid_size_px=(3, 11), name="AllChannelsCLAHE"), iaa.CLAHE(clip_limit=(2, 10), tile_grid_size_px=(3, 11), to_colorspace="HSV", name="CLAHE"), iaa.AllChannelsHistogramEqualization( name="AllChannelsHistogramEqualization"), iaa.HistogramEqualization(to_colorspace="HSV", name="HistogramEqualization"), ] augmenters_convolutional = [ iaa.Convolve(np.float32([[0, 0, 0], [0, 1, 0], [0, 0, 0]]), name="Convolve_3x3"), iaa.Sharpen(alpha=(0.01, 0.99), lightness=(0.5, 2), name="Sharpen"), iaa.Emboss(alpha=(0.01, 0.99), strength=(0, 2), name="Emboss"), iaa.EdgeDetect(alpha=(0.01, 0.99), name="EdgeDetect"), iaa.DirectedEdgeDetect(alpha=(0.01, 0.99), name="DirectedEdgeDetect") ] augmenters_edges = [iaa.Canny(alpha=(0.01, 0.99), name="Canny")] augmenters_flip = [ iaa.Fliplr(1.0, name="Fliplr"), iaa.Flipud(1.0, name="Flipud") ] augmenters_geometric = [ iaa.Affine(scale=(0.9, 1.1), translate_percent={ "x": (-0.05, 0.05), "y": (-0.05, 0.05) }, rotate=(-10, 10), shear=(-10, 10), order=0, mode="constant", cval=(0, 255), name="Affine_order_0_constant"), iaa.Affine(scale=(0.9, 1.1), translate_percent={ "x": (-0.05, 0.05), "y": (-0.05, 0.05) }, rotate=(-10, 10), shear=(-10, 10), order=1, mode="constant", cval=(0, 255), name="Affine_order_1_constant"), iaa.Affine(scale=(0.9, 1.1), translate_percent={ "x": (-0.05, 0.05), "y": (-0.05, 0.05) }, rotate=(-10, 10), shear=(-10, 10), order=3, mode="constant", cval=(0, 255), name="Affine_order_3_constant"), iaa.Affine(scale=(0.9, 1.1), translate_percent={ "x": (-0.05, 0.05), "y": (-0.05, 0.05) }, rotate=(-10, 10), shear=(-10, 10), order=1, mode="edge", cval=(0, 255), name="Affine_order_1_edge"), iaa.Affine(scale=(0.9, 1.1), translate_percent={ "x": (-0.05, 0.05), "y": (-0.05, 0.05) }, rotate=(-10, 10), shear=(-10, 10), order=1, mode="constant", cval=(0, 255), backend="skimage", name="Affine_order_1_constant_skimage"), # TODO AffineCv2 iaa.PiecewiseAffine(scale=(0.01, 0.05), nb_rows=4, nb_cols=4, order=1, mode="constant", name="PiecewiseAffine_4x4_order_1_constant"), iaa.PiecewiseAffine(scale=(0.01, 0.05), nb_rows=4, nb_cols=4, order=0, mode="constant", name="PiecewiseAffine_4x4_order_0_constant"), iaa.PiecewiseAffine(scale=(0.01, 0.05), nb_rows=4, nb_cols=4, order=1, mode="edge", name="PiecewiseAffine_4x4_order_1_edge"), iaa.PiecewiseAffine(scale=(0.01, 0.05), nb_rows=8, nb_cols=8, order=1, mode="constant", name="PiecewiseAffine_8x8_order_1_constant"), iaa.PerspectiveTransform(scale=(0.01, 0.05), keep_size=False, name="PerspectiveTransform"), iaa.PerspectiveTransform(scale=(0.01, 0.05), keep_size=True, name="PerspectiveTransform_keep_size"), iaa.ElasticTransformation( alpha=(1, 10), sigma=(0.5, 1.5), order=0, mode="constant", cval=0, name="ElasticTransformation_order_0_constant"), iaa.ElasticTransformation( alpha=(1, 10), sigma=(0.5, 1.5), order=1, mode="constant", cval=0, name="ElasticTransformation_order_1_constant"), iaa.ElasticTransformation( alpha=(1, 10), sigma=(0.5, 1.5), order=1, mode="nearest", cval=0, name="ElasticTransformation_order_1_nearest"), iaa.ElasticTransformation( alpha=(1, 10), sigma=(0.5, 1.5), order=1, mode="reflect", cval=0, name="ElasticTransformation_order_1_reflect"), iaa.Rot90((1, 3), keep_size=False, name="Rot90"), iaa.Rot90((1, 3), keep_size=True, name="Rot90_keep_size") ] augmenters_pooling = [ iaa.AveragePooling(kernel_size=(1, 16), keep_size=False, name="AveragePooling"), iaa.AveragePooling(kernel_size=(1, 16), keep_size=True, name="AveragePooling_keep_size"), iaa.MaxPooling(kernel_size=(1, 16), keep_size=False, name="MaxPooling"), iaa.MaxPooling(kernel_size=(1, 16), keep_size=True, name="MaxPooling_keep_size"), iaa.MinPooling(kernel_size=(1, 16), keep_size=False, name="MinPooling"), iaa.MinPooling(kernel_size=(1, 16), keep_size=True, name="MinPooling_keep_size"), iaa.MedianPooling(kernel_size=(1, 16), keep_size=False, name="MedianPooling"), iaa.MedianPooling(kernel_size=(1, 16), keep_size=True, name="MedianPooling_keep_size") ] augmenters_segmentation = [ iaa.Superpixels(p_replace=(0.05, 1.0), n_segments=(10, 100), max_size=64, interpolation="cubic", name="Superpixels_max_size_64_cubic"), iaa.Superpixels(p_replace=(0.05, 1.0), n_segments=(10, 100), max_size=64, interpolation="linear", name="Superpixels_max_size_64_linear"), iaa.Superpixels(p_replace=(0.05, 1.0), n_segments=(10, 100), max_size=128, interpolation="linear", name="Superpixels_max_size_128_linear"), iaa.Superpixels(p_replace=(0.05, 1.0), n_segments=(10, 100), max_size=224, interpolation="linear", name="Superpixels_max_size_224_linear"), iaa.UniformVoronoi(n_points=(250, 1000), name="UniformVoronoi"), iaa.RegularGridVoronoi(n_rows=(16, 31), n_cols=(16, 31), name="RegularGridVoronoi"), iaa.RelativeRegularGridVoronoi(n_rows_frac=(0.07, 0.14), n_cols_frac=(0.07, 0.14), name="RelativeRegularGridVoronoi"), ] augmenters_size = [ iaa.Resize((0.8, 1.2), interpolation="nearest", name="Resize_nearest"), iaa.Resize((0.8, 1.2), interpolation="linear", name="Resize_linear"), iaa.Resize((0.8, 1.2), interpolation="cubic", name="Resize_cubic"), iaa.CropAndPad(percent=(-0.2, 0.2), pad_mode="constant", pad_cval=(0, 255), keep_size=False, name="CropAndPad"), iaa.CropAndPad(percent=(-0.2, 0.2), pad_mode="edge", pad_cval=(0, 255), keep_size=False, name="CropAndPad_edge"), iaa.CropAndPad(percent=(-0.2, 0.2), pad_mode="constant", pad_cval=(0, 255), name="CropAndPad_keep_size"), iaa.Pad(percent=(0.05, 0.2), pad_mode="constant", pad_cval=(0, 255), keep_size=False, name="Pad"), iaa.Pad(percent=(0.05, 0.2), pad_mode="edge", pad_cval=(0, 255), keep_size=False, name="Pad_edge"), iaa.Pad(percent=(0.05, 0.2), pad_mode="constant", pad_cval=(0, 255), name="Pad_keep_size"), iaa.Crop(percent=(0.05, 0.2), keep_size=False, name="Crop"), iaa.Crop(percent=(0.05, 0.2), name="Crop_keep_size"), iaa.PadToFixedSize(width=width + 10, height=height + 10, pad_mode="constant", pad_cval=(0, 255), name="PadToFixedSize"), iaa.CropToFixedSize(width=width - 10, height=height - 10, name="CropToFixedSize"), iaa.KeepSizeByResize(iaa.CropToFixedSize(height=height - 10, width=width - 10), interpolation="nearest", name="KeepSizeByResize_CropToFixedSize_nearest"), iaa.KeepSizeByResize(iaa.CropToFixedSize(height=height - 10, width=width - 10), interpolation="linear", name="KeepSizeByResize_CropToFixedSize_linear"), iaa.KeepSizeByResize(iaa.CropToFixedSize(height=height - 10, width=width - 10), interpolation="cubic", name="KeepSizeByResize_CropToFixedSize_cubic"), ] augmenters_weather = [ iaa.FastSnowyLandscape(lightness_threshold=(100, 255), lightness_multiplier=(1.0, 4.0), name="FastSnowyLandscape"), iaa.Clouds(name="Clouds"), iaa.Fog(name="Fog"), iaa.CloudLayer(intensity_mean=(196, 255), intensity_freq_exponent=(-2.5, -2.0), intensity_coarse_scale=10, alpha_min=0, alpha_multiplier=(0.25, 0.75), alpha_size_px_max=(2, 8), alpha_freq_exponent=(-2.5, -2.0), sparsity=(0.8, 1.0), density_multiplier=(0.5, 1.0), name="CloudLayer"), iaa.Snowflakes(name="Snowflakes"), iaa.SnowflakesLayer(density=(0.005, 0.075), density_uniformity=(0.3, 0.9), flake_size=(0.2, 0.7), flake_size_uniformity=(0.4, 0.8), angle=(-30, 30), speed=(0.007, 0.03), blur_sigma_fraction=(0.0001, 0.001), name="SnowflakesLayer") ] augmenters = (augmenters_meta + augmenters_arithmetic + augmenters_blend + augmenters_blur + augmenters_color + augmenters_contrast + augmenters_convolutional + augmenters_edges + augmenters_flip + augmenters_geometric + augmenters_pooling + augmenters_segmentation + augmenters_size + augmenters_weather) if only_augmenters is not None: augmenters_reduced = [] for augmenter in augmenters: if any([ re.search(pattern, augmenter.name) for pattern in only_augmenters ]): augmenters_reduced.append(augmenter) augmenters = augmenters_reduced return augmenters
augmenters = [ iaa.GaussianBlur(sigma=(1.0, 3.0)), # blur images with a sigma of 0 to 2.0 iaa.MotionBlur((3, 5)), # blur image iaa.AdditiveGaussianNoise( scale=(0, 0.05 * 255) ), # Add gaussian noise to an image, sampled once per pixel from a normal distribution N(0, 0.05*255) iaa.AdditiveLaplaceNoise(scale=(0, 0.05 * 255)), iaa.Multiply((0.5, 1.5), per_channel=0.5), iaa.Multiply( (0.5, 1.5)), # Multiply each image with a random value between 0.5 and 1.5: iaa.Dropout( p=(0.05, 0.15) ), # Sample per image a value p from the range 0.05<=p<=0.15 and then drop p percent of all pixels in the image (i.e. convert them to black pixels): iaa.ImpulseNoise(p=(0.03, 0.06)), iaa.Salt(p=(0.03, 0.05)), iaa.Add((-30, 30)), # adding random value to pixels iaa.SigmoidContrast(6.1, 0.5), iaa.PerspectiveTransform(scale=0.02), iaa.PerspectiveTransform(scale=0.01), iaa.PerspectiveTransform(scale=0.015), iaa.PerspectiveTransform(scale=0.012), iaa.PiecewiseAffine(scale=0.015), iaa.PiecewiseAffine(scale=0.005), iaa.PiecewiseAffine(scale=0.01), iaa.Crop(px=(10, 15), keep_size=True), iaa.Crop(px=(10, 15)) ] def check_dir_or_create(dir):
def train(tip, iters=None, learning_rate=0.001, batch_norm=False): import tensorflow as tf random.seed(datetime.datetime.now()) tf.set_random_seed(seed()) default_device = '/gpu:0' # default_device = '/cpu:0' # hyperparams batch_size = 94 training = True batch_norms_training = False # steps light_summary_steps = 25 heavy_summary_steps = 250 checkpoint_steps = 500 # stats / logging model_version = 1 time_str = datetime.datetime.now().strftime('%m-%d--%H%M%S') best_loss = 0.6 batch_num = 0 best_acc = 0 models_to_keep = 3 # glob vars heavy_sum = [] light_sum = [] models_history = [] train_only_dense = False dense_size = 64 dropout = True learning_rate = 1e-6 augmentation = True with tf.device(default_device): # config = tf.ConfigProto() config = tf.ConfigProto(log_device_placement=False, allow_soft_placement=True) # please do not use the totality of the GPU memory config.gpu_options.per_process_gpu_memory_fraction = 0.98 # config = tf.ConfigProto(device_count = {'GPU': 0}) config.gpu_options.allow_growth = True with tf.Session(graph=tf.Graph(), config=config) as sess: with tf.name_scope("inputs"): # _images = tf.placeholder(tf.float32, [None, 224, 224, 1]) _images = tf.placeholder(tf.float32, [batch_size, 224, 224, 1]) _is_training = tf.placeholder(tf.bool, name='is_training') model = vggish.Vggish(_images, classes=2, trainable=training, batch_norm=batch_norm, dropout=dropout, only_dense=train_only_dense, dense_size=dense_size, bn_trainable=batch_norms_training) with tf.name_scope("targets"): _labels = tf.placeholder(tf.float32, shape=(None, 2), name='labels') with tf.name_scope("outputs"): logits = model.fc3l # predictions = tf.nn.softmax(logits, name='predictions') predictions = tf.nn.sigmoid(logits, name='predictions') tvars = tf.trainable_variables() for v in tvars: print(v) if 'weights' in v.name: heavy_sum.append(tf.summary.histogram(v.name, v)) # if 'conv1_1' in v.name: # light_sum.append(tf.summary.histogram(v.name, v)) for v in tvars: if 'bias' in v.name: heavy_sum.append(tf.summary.histogram(v.name, v)) # if 'conv1_1' in v.name: # light_sum.append(tf.summary.histogram(v.name, v)) light_sum.append(tf.summary.histogram("predictions", predictions)) light_sum.extend(model.summaries) heavy_sum.extend(model.heavy_summaries) with tf.name_scope("0_cost"): cross_entropy = tf.nn.sigmoid_cross_entropy_with_logits( logits=logits, labels=_labels, name='cross_entropy' ) tvars = tf.trainable_variables() L2 = [tf.nn.l2_loss(v) for v in tvars if 'bias' not in v.name] lossL2 = tf.add_n(L2) * 0.01 cost = tf.reduce_mean(cross_entropy, name='cost') + lossL2 light_sum.append(tf.summary.scalar("cost", cost)) def my_capper(t): print(t) # return t if t is None: return None return tf.clip_by_value(t, -5., 5.) log_string = 'logs/{}-vggish/{}-lr-{:.8f}'.format(model_version, time_str, learning_rate) with tf.name_scope("0_train"): with tf.control_dependencies(tf.get_collection(tf.GraphKeys.UPDATE_OPS)): vars_to_train = model.get_vars_to_train() global_step = tf.Variable(0, name='global_step') learning_rate = tf.train.exponential_decay(learning_rate, global_step, 100000, 0.96, staircase=True) optimizer = tf.train.AdamOptimizer(learning_rate=learning_rate) # optimizer = tf.train.GradientDescentOptimizer(learning_rate=learning_rate) grads_and_vars = optimizer.compute_gradients(cost, var_list=vars_to_train) grads_and_vars = [(my_capper(gv[0]), gv[1]) for gv in grads_and_vars] optimizer = optimizer.apply_gradients(grads_and_vars, global_step=global_step) correct_predictions = tf.equal(tf.argmax(predictions, 1), tf.argmax(_labels, 1), name='correct_predictions') accuracy = tf.reduce_mean(tf.cast(correct_predictions, tf.float32), name='accuracy') grad_norm = tf.norm(grads_and_vars[0][0]) light_sum.append(tf.summary.scalar("accuracy", accuracy)) light_sum.append(tf.summary.scalar("gradient", grad_norm)) light_summary = tf.summary.merge(light_sum) heavy_summary = tf.summary.merge(light_sum + heavy_sum) sess.run(tf.global_variables_initializer()) saver = tf.train.Saver(tf.global_variables(), max_to_keep=None) iteration = 0 try: print('[+] loading startup.json') startup = json.load(open('startup.vggish.json', 'r')) print('[+] loading path:', startup['path']) state = json.load(open(startup['path'], 'r')) print('[+] loading checkpoint:', state['checkpoint_path']) last_checkpoint = os.path.dirname(state['checkpoint_path']) weights_to_load = vggish.conv_vars + vggish.fc_vars if train_only_dense: weights_to_load = vggish.conv_vars model.load_weights(last_checkpoint, vars_names=weights_to_load) iteration = state['iteration'] best_loss = state['best_loss'] if 'train_loss' in state: best_loss = state['best_loss'] checkpoint_path = state['checkpoint_path'] except: print('[!] no models to checkpoint from..') raise writer = tf.summary.FileWriter(log_string) augmenters = [ # blur images with a sigma between 0 and 3.0 iaa.Noop(), iaa.GaussianBlur(sigma=(0.5, 2.0)), iaa.Add((-50.0, 50.0), per_channel=False), iaa.AdditiveGaussianNoise(loc=0, scale=(0.07*255, 0.07*255), per_channel=False), iaa.Dropout(p=0.07, per_channel=False), iaa.CoarseDropout(p=(0.05, 0.15), size_percent=(0.1, 0.9), per_channel=False), iaa.SaltAndPepper(p=(0.05, 0.15), per_channel=False), iaa.Salt(p=(0.05, 0.15), per_channel=False), iaa.Pepper(p=(0.05, 0.15), per_channel=False), iaa.ContrastNormalization(alpha=(iap.Uniform(0.02, 0.03), iap.Uniform(1.7, 2.1))), iaa.ElasticTransformation(alpha=(0.5, 2.0)), ] if not augmentation: augmenters = [ iaa.Noop(), ] seq = iaa.Sequential( iaa.OneOf(augmenters), ) while True: if iters and iteration >= iters: return seq.reseed(seed()) np.random.seed(seed()) descriptions, features, labels = get_data_from_tip(tip, batch_size) iteration += 1 # import pdb; pdb.set_trace() mean = 0.172840994091 std = 0.206961060284 merged_summaries = light_summary if iteration % heavy_summary_steps == 0: merged_summaries = heavy_summary try: with np.errstate(all='raise'): for i in range(5): newfeatures = seq.augment_images(features * 255) / 255 if not np.isnan(newfeatures).any(): break print('[!] has nan in newfeatures, retrying', i) if np.isnan(newfeatures).any(): print('[!] could not get rid of nan.. skipping this batch') iteration -= 1 continue features = newfeatures except Exception: print("[!] Warning detected augmenting, skipping..") tb = traceback.format_exc() open("numpy_warns.log", 'ab').write(str(descriptions).encode('utf-8')) open("numpy_warns.log", 'ab').write(str(tb).encode('utf-8')) open("numpy_warns.log", 'a').write('------------------------------------') # import pdb; pdb.set_trace() continue feed_dict = { _images: features, _labels: labels, _is_training: training } train_loss, val_acc, _, p, summary, grad, _corr_pred = sess.run( [cost, accuracy, optimizer, logits, merged_summaries, grad_norm, correct_predictions], feed_dict=feed_dict ) print('[+] iteration {}'.format(iteration)) if iteration % light_summary_steps == 0: if os.path.isfile('nomix_pdb'): import pdb pdb.set_trace() print('[+] writing summary') writer.add_summary(summary, iteration) print('\tIteration {} Accuracy: {} Loss: {}/{}'.format(iteration, val_acc, train_loss, best_loss)) # if train_loss < best_loss: if iteration % checkpoint_steps == 0: # print('\t\tNew Best Loss!') best_loss = train_loss timestamp = datetime.datetime.now().strftime("%Y-%m-%d_%H%M%S") checkpoint_dir = os.path.join('D:\\checkpoint', timestamp) checkpoint_path = os.path.join('D:\\checkpoint', timestamp, 'model.ckpt') print('\t\tSaving model to:' + checkpoint_path) saver.save(sess, checkpoint_path, global_step=batch_num) state = { 'iteration': iteration, 'best_acc': float(best_acc), 'best_loss': float(best_loss), 'val_acc': float(val_acc), 'train_loss': float(train_loss), 'checkpoint_path': checkpoint_path, 'log_string': log_string, } # state_path = os.path.join('save', timestamp, 'state.json') state_path = os.path.join('D:\\checkpoint', timestamp, 'state.json') open(state_path, 'w').write(json.dumps(state)) startup = { 'path': state_path, } open('startup.vggish.json', 'w').write(json.dumps(startup)) models_history.append(checkpoint_dir) while len(models_history) > models_to_keep: try: path_to_del = models_history.pop(0) print('[+] deleting model', path_to_del) shutil.rmtree(path_to_del) except: print('[+] failed to delete') traceback.print_exc()
def draw_per_augmenter_images(): print("[draw_per_augmenter_images] Loading image...") #image = misc.imresize(ndimage.imread("quokka.jpg")[0:643, 0:643], (128, 128)) image = ia.quokka_square(size=(128, 128)) keypoints = [ia.Keypoint(x=34, y=15), ia.Keypoint(x=85, y=13), ia.Keypoint(x=63, y=73)] # left ear, right ear, mouth keypoints = [ia.KeypointsOnImage(keypoints, shape=image.shape)] print("[draw_per_augmenter_images] Initializing...") rows_augmenters = [ (0, "Noop", [("", iaa.Noop()) for _ in sm.xrange(5)]), (0, "Crop\n(top, right,\nbottom, left)", [(str(vals), iaa.Crop(px=vals)) for vals in [(2, 0, 0, 0), (0, 8, 8, 0), (4, 0, 16, 4), (8, 0, 0, 32), (32, 64, 0, 0)]]), (0, "Pad\n(top, right,\nbottom, left)", [(str(vals), iaa.Pad(px=vals)) for vals in [(2, 0, 0, 0), (0, 8, 8, 0), (4, 0, 16, 4), (8, 0, 0, 32), (32, 64, 0, 0)]]), (0, "Fliplr", [(str(p), iaa.Fliplr(p)) for p in [0, 0, 1, 1, 1]]), (0, "Flipud", [(str(p), iaa.Flipud(p)) for p in [0, 0, 1, 1, 1]]), (0, "Superpixels\np_replace=1", [("n_segments=%d" % (n_segments,), iaa.Superpixels(p_replace=1.0, n_segments=n_segments)) for n_segments in [25, 50, 75, 100, 125]]), (0, "Superpixels\nn_segments=100", [("p_replace=%.2f" % (p_replace,), iaa.Superpixels(p_replace=p_replace, n_segments=100)) for p_replace in [0, 0.25, 0.5, 0.75, 1.0]]), (0, "Invert", [("p=%d" % (p,), iaa.Invert(p=p)) for p in [0, 0, 1, 1, 1]]), (0, "Invert\n(per_channel)", [("p=%.2f" % (p,), iaa.Invert(p=p, per_channel=True)) for p in [0.5, 0.5, 0.5, 0.5, 0.5]]), (0, "Add", [("value=%d" % (val,), iaa.Add(val)) for val in [-45, -25, 0, 25, 45]]), (0, "Add\n(per channel)", [("value=(%d, %d)" % (vals[0], vals[1],), iaa.Add(vals, per_channel=True)) for vals in [(-55, -35), (-35, -15), (-10, 10), (15, 35), (35, 55)]]), (0, "AddToHueAndSaturation", [("value=%d" % (val,), iaa.AddToHueAndSaturation(val)) for val in [-45, -25, 0, 25, 45]]), (0, "Multiply", [("value=%.2f" % (val,), iaa.Multiply(val)) for val in [0.25, 0.5, 1.0, 1.25, 1.5]]), (1, "Multiply\n(per channel)", [("value=(%.2f, %.2f)" % (vals[0], vals[1],), iaa.Multiply(vals, per_channel=True)) for vals in [(0.15, 0.35), (0.4, 0.6), (0.9, 1.1), (1.15, 1.35), (1.4, 1.6)]]), (0, "GaussianBlur", [("sigma=%.2f" % (sigma,), iaa.GaussianBlur(sigma=sigma)) for sigma in [0.25, 0.50, 1.0, 2.0, 4.0]]), (0, "AverageBlur", [("k=%d" % (k,), iaa.AverageBlur(k=k)) for k in [1, 3, 5, 7, 9]]), (0, "MedianBlur", [("k=%d" % (k,), iaa.MedianBlur(k=k)) for k in [1, 3, 5, 7, 9]]), (0, "BilateralBlur\nsigma_color=250,\nsigma_space=250", [("d=%d" % (d,), iaa.BilateralBlur(d=d, sigma_color=250, sigma_space=250)) for d in [1, 3, 5, 7, 9]]), (0, "Sharpen\n(alpha=1)", [("lightness=%.2f" % (lightness,), iaa.Sharpen(alpha=1, lightness=lightness)) for lightness in [0, 0.5, 1.0, 1.5, 2.0]]), (0, "Emboss\n(alpha=1)", [("strength=%.2f" % (strength,), iaa.Emboss(alpha=1, strength=strength)) for strength in [0, 0.5, 1.0, 1.5, 2.0]]), (0, "EdgeDetect", [("alpha=%.2f" % (alpha,), iaa.EdgeDetect(alpha=alpha)) for alpha in [0.0, 0.25, 0.5, 0.75, 1.0]]), (0, "DirectedEdgeDetect\n(alpha=1)", [("direction=%.2f" % (direction,), iaa.DirectedEdgeDetect(alpha=1, direction=direction)) for direction in [0.0, 1*(360/5)/360, 2*(360/5)/360, 3*(360/5)/360, 4*(360/5)/360]]), (0, "AdditiveGaussianNoise", [("scale=%.2f*255" % (scale,), iaa.AdditiveGaussianNoise(scale=scale * 255)) for scale in [0.025, 0.05, 0.1, 0.2, 0.3]]), (0, "AdditiveGaussianNoise\n(per channel)", [("scale=%.2f*255" % (scale,), iaa.AdditiveGaussianNoise(scale=scale * 255, per_channel=True)) for scale in [0.025, 0.05, 0.1, 0.2, 0.3]]), (0, "Dropout", [("p=%.2f" % (p,), iaa.Dropout(p=p)) for p in [0.025, 0.05, 0.1, 0.2, 0.4]]), (0, "Dropout\n(per channel)", [("p=%.2f" % (p,), iaa.Dropout(p=p, per_channel=True)) for p in [0.025, 0.05, 0.1, 0.2, 0.4]]), (3, "CoarseDropout\n(p=0.2)", [("size_percent=%.2f" % (size_percent,), iaa.CoarseDropout(p=0.2, size_percent=size_percent, min_size=2)) for size_percent in [0.3, 0.2, 0.1, 0.05, 0.02]]), (0, "CoarseDropout\n(p=0.2, per channel)", [("size_percent=%.2f" % (size_percent,), iaa.CoarseDropout(p=0.2, size_percent=size_percent, per_channel=True, min_size=2)) for size_percent in [0.3, 0.2, 0.1, 0.05, 0.02]]), (0, "SaltAndPepper", [("p=%.2f" % (p,), iaa.SaltAndPepper(p=p)) for p in [0.025, 0.05, 0.1, 0.2, 0.4]]), (0, "Salt", [("p=%.2f" % (p,), iaa.Salt(p=p)) for p in [0.025, 0.05, 0.1, 0.2, 0.4]]), (0, "Pepper", [("p=%.2f" % (p,), iaa.Pepper(p=p)) for p in [0.025, 0.05, 0.1, 0.2, 0.4]]), (0, "CoarseSaltAndPepper\n(p=0.2)", [("size_percent=%.2f" % (size_percent,), iaa.CoarseSaltAndPepper(p=0.2, size_percent=size_percent, min_size=2)) for size_percent in [0.3, 0.2, 0.1, 0.05, 0.02]]), (0, "CoarseSalt\n(p=0.2)", [("size_percent=%.2f" % (size_percent,), iaa.CoarseSalt(p=0.2, size_percent=size_percent, min_size=2)) for size_percent in [0.3, 0.2, 0.1, 0.05, 0.02]]), (0, "CoarsePepper\n(p=0.2)", [("size_percent=%.2f" % (size_percent,), iaa.CoarsePepper(p=0.2, size_percent=size_percent, min_size=2)) for size_percent in [0.3, 0.2, 0.1, 0.05, 0.02]]), (0, "ContrastNormalization", [("alpha=%.1f" % (alpha,), iaa.ContrastNormalization(alpha=alpha)) for alpha in [0.5, 0.75, 1.0, 1.25, 1.50]]), (0, "ContrastNormalization\n(per channel)", [("alpha=(%.2f, %.2f)" % (alphas[0], alphas[1],), iaa.ContrastNormalization(alpha=alphas, per_channel=True)) for alphas in [(0.4, 0.6), (0.65, 0.85), (0.9, 1.1), (1.15, 1.35), (1.4, 1.6)]]), (0, "Grayscale", [("alpha=%.1f" % (alpha,), iaa.Grayscale(alpha=alpha)) for alpha in [0.0, 0.25, 0.5, 0.75, 1.0]]), (6, "PerspectiveTransform", [("scale=%.3f" % (scale,), iaa.PerspectiveTransform(scale=scale)) for scale in [0.025, 0.05, 0.075, 0.10, 0.125]]), (0, "PiecewiseAffine", [("scale=%.3f" % (scale,), iaa.PiecewiseAffine(scale=scale)) for scale in [0.015, 0.03, 0.045, 0.06, 0.075]]), (0, "Affine: Scale", [("%.1fx" % (scale,), iaa.Affine(scale=scale)) for scale in [0.1, 0.5, 1.0, 1.5, 1.9]]), (0, "Affine: Translate", [("x=%d y=%d" % (x, y), iaa.Affine(translate_px={"x": x, "y": y})) for x, y in [(-32, -16), (-16, -32), (-16, -8), (16, 8), (16, 32)]]), (0, "Affine: Rotate", [("%d deg" % (rotate,), iaa.Affine(rotate=rotate)) for rotate in [-90, -45, 0, 45, 90]]), (0, "Affine: Shear", [("%d deg" % (shear,), iaa.Affine(shear=shear)) for shear in [-45, -25, 0, 25, 45]]), (0, "Affine: Modes", [(mode, iaa.Affine(translate_px=-32, mode=mode)) for mode in ["constant", "edge", "symmetric", "reflect", "wrap"]]), (0, "Affine: cval", [("%d" % (int(cval*255),), iaa.Affine(translate_px=-32, cval=int(cval*255), mode="constant")) for cval in [0.0, 0.25, 0.5, 0.75, 1.0]]), ( 2, "Affine: all", [ ( "", iaa.Affine( scale={"x": (0.5, 1.5), "y": (0.5, 1.5)}, translate_px={"x": (-32, 32), "y": (-32, 32)}, rotate=(-45, 45), shear=(-32, 32), mode=ia.ALL, cval=(0.0, 1.0) ) ) for _ in sm.xrange(5) ] ), (1, "ElasticTransformation\n(sigma=0.2)", [("alpha=%.1f" % (alpha,), iaa.ElasticTransformation(alpha=alpha, sigma=0.2)) for alpha in [0.1, 0.5, 1.0, 3.0, 9.0]]), (0, "Alpha\nwith EdgeDetect(1.0)", [("factor=%.1f" % (factor,), iaa.Alpha(factor=factor, first=iaa.EdgeDetect(1.0))) for factor in [0.0, 0.25, 0.5, 0.75, 1.0]]), (4, "Alpha\nwith EdgeDetect(1.0)\n(per channel)", [("factor=(%.2f, %.2f)" % (factor[0], factor[1]), iaa.Alpha(factor=factor, first=iaa.EdgeDetect(1.0), per_channel=0.5)) for factor in [(0.0, 0.2), (0.15, 0.35), (0.4, 0.6), (0.65, 0.85), (0.8, 1.0)]]), (15, "SimplexNoiseAlpha\nwith EdgeDetect(1.0)", [("", iaa.SimplexNoiseAlpha(first=iaa.EdgeDetect(1.0))) for alpha in [0.0, 0.25, 0.5, 0.75, 1.0]]), (9, "FrequencyNoiseAlpha\nwith EdgeDetect(1.0)", [("exponent=%.1f" % (exponent,), iaa.FrequencyNoiseAlpha(exponent=exponent, first=iaa.EdgeDetect(1.0), size_px_max=16, upscale_method="linear", sigmoid=False)) for exponent in [-4, -2, 0, 2, 4]]) ] print("[draw_per_augmenter_images] Augmenting...") rows = [] for (row_seed, row_name, augmenters) in rows_augmenters: ia.seed(row_seed) #for img_title, augmenter in augmenters: # #aug.reseed(1000) # pass row_images = [] row_keypoints = [] row_titles = [] for img_title, augmenter in augmenters: aug_det = augmenter.to_deterministic() row_images.append(aug_det.augment_image(image)) row_keypoints.append(aug_det.augment_keypoints(keypoints)[0]) row_titles.append(img_title) rows.append((row_name, row_images, row_keypoints, row_titles)) # matplotlib drawin routine """ print("[draw_per_augmenter_images] Plotting...") width = 8 height = int(1.5 * len(rows_augmenters)) fig = plt.figure(figsize=(width, height)) grid_rows = len(rows) grid_cols = 1 + 5 gs = gridspec.GridSpec(grid_rows, grid_cols, width_ratios=[2, 1, 1, 1, 1, 1]) axes = [] for i in sm.xrange(grid_rows): axes.append([plt.subplot(gs[i, col_idx]) for col_idx in sm.xrange(grid_cols)]) fig.tight_layout() #fig.subplots_adjust(bottom=0.2 / grid_rows, hspace=0.22) #fig.subplots_adjust(wspace=0.005, hspace=0.425, bottom=0.02) fig.subplots_adjust(wspace=0.005, hspace=0.005, bottom=0.02) for row_idx, (row_name, row_images, row_keypoints, row_titles) in enumerate(rows): axes_row = axes[row_idx] for col_idx in sm.xrange(grid_cols): ax = axes_row[col_idx] ax.cla() ax.axis("off") ax.get_xaxis().set_visible(False) ax.get_yaxis().set_visible(False) if col_idx == 0: ax.text(0, 0.5, row_name, color="black") else: cell_image = row_images[col_idx-1] cell_keypoints = row_keypoints[col_idx-1] cell_image_kp = cell_keypoints.draw_on_image(cell_image, size=5) ax.imshow(cell_image_kp) x = 0 y = 145 #ax.text(x, y, row_titles[col_idx-1], color="black", backgroundcolor="white", fontsize=6) ax.text(x, y, row_titles[col_idx-1], color="black", fontsize=7) fig.savefig("examples.jpg", bbox_inches="tight") #plt.show() """ # simpler and faster drawing routine """ output_image = ExamplesImage(128, 128, 128+64, 32) for (row_name, row_images, row_keypoints, row_titles) in rows: row_images_kps = [] for image, keypoints in zip(row_images, row_keypoints): row_images_kps.append(keypoints.draw_on_image(image, size=5)) output_image.add_row(row_name, row_images_kps, row_titles) misc.imsave("examples.jpg", output_image.draw()) """ # routine to draw many single files seen = defaultdict(lambda: 0) markups = [] for (row_name, row_images, row_keypoints, row_titles) in rows: output_image = ExamplesImage(128, 128, 128+64, 32) row_images_kps = [] for image, keypoints in zip(row_images, row_keypoints): row_images_kps.append(keypoints.draw_on_image(image, size=5)) output_image.add_row(row_name, row_images_kps, row_titles) if "\n" in row_name: row_name_clean = row_name[0:row_name.find("\n")+1] else: row_name_clean = row_name row_name_clean = re.sub(r"[^a-z0-9]+", "_", row_name_clean.lower()) row_name_clean = row_name_clean.strip("_") if seen[row_name_clean] > 0: row_name_clean = "%s_%d" % (row_name_clean, seen[row_name_clean] + 1) fp = os.path.join(IMAGES_DIR, "examples_%s.jpg" % (row_name_clean,)) #misc.imsave(fp, output_image.draw()) save(fp, output_image.draw()) seen[row_name_clean] += 1 markup_descr = row_name.replace('"', '') \ .replace("\n", " ") \ .replace("(", "") \ .replace(")", "") markup = '![%s](%s?raw=true "%s")' % (markup_descr, fp, markup_descr) markups.append(markup) for markup in markups: print(markup)
def arithmetic(image_array: ndarray): seq = iaa.Sequential([ iaa.Salt(p=0.03) ]) return seq.augment_image(image_array)