def __init__(self, root_dir, transform=None):
samples = parse_car_csv(root_dir)
jpeg_images = [
cv2.imread(
os.path.join(root_dir, 'JPEGImages', sample_name + ".jpg"))
for sample_name in samples
]
positive_annotations = [
os.path.join(root_dir, 'Annotations', sample_name + '_1.csv')
for sample_name in samples
]
negative_annotations = [
os.path.join(root_dir, 'Annotations', sample_name + '_0.csv')
for sample_name in samples
]
# 边界框大小
positive_sizes = list()
negative_sizes = list()
# 边界框坐标
positive_rects = list()
negative_rects = list()
for annotation_path in positive_annotations:
rects = np.loadtxt(annotation_path, dtype=np.int, delimiter=' ')
# 存在文件为空或者文件中仅有单行数据
if len(rects.shape) == 1:
# 是否为单行
if rects.shape[0] == 4:
positive_rects.append(rects)
positive_sizes.append(1)
else:
positive_sizes.append(0)
else:
positive_rects.extend(rects)
positive_sizes.append(len(rects))
for annotation_path in negative_annotations:
rects = np.loadtxt(annotation_path, dtype=np.int, delimiter=' ')
# 和正样本规则一样
if len(rects.shape) == 1:
if rects.shape[0] == 4:
negative_rects.append(rects)
negative_sizes.append(1)
else:
positive_sizes.append(0)
else:
negative_rects.extend(rects)
negative_sizes.append(len(rects))
self.transform = transform
self.jpeg_images = jpeg_images
self.positive_sizes = positive_sizes
self.negative_sizes = negative_sizes
self.positive_rects = positive_rects
self.negative_rects = negative_rects
self.total_positive_num = int(np.sum(positive_sizes))
self.total_negative_num = int(np.sum(negative_sizes))
def __init__(self, root_dir, transform):
"""
加载所有的图像以及正负样本边界框
"""
self.transform = transform
samples = util.parse_car_csv(root_dir)
jpeg_images = list()
annotation_dict = dict()
for idx in range(len(samples)):
sample_name = samples[idx]
img = cv2.imread(
os.path.join(root_dir, 'JPEGImages', sample_name + ".jpg"))
h, w = img.shape[:2]
jpeg_images.append(img)
positive_annotation_path = os.path.join(root_dir, 'Annotations',
sample_name + '_1.csv')
positive_annotations = np.loadtxt(positive_annotation_path,
dtype=np.float,
delimiter=' ')
if len(positive_annotations.shape) == 1:
positive_annotations = positive_annotations[np.newaxis, :]
# print(positive_annotations.shape)
positive_annotations[:, 0] /= w
positive_annotations[:, 1] /= h
positive_annotations[:, 2] /= w
positive_annotations[:, 3] /= h
# if len(positive_annotations) < 16:
# print(sample_name)
negative_annotation_path = os.path.join(root_dir, 'Annotations',
sample_name + '_0.csv')
negative_annotations = np.loadtxt(negative_annotation_path,
dtype=np.float,
delimiter=' ')
negative_annotations[:, 0] /= w
negative_annotations[:, 1] /= h
negative_annotations[:, 2] /= w
negative_annotations[:, 3] /= h
annotation_dict[str(idx)] = {
'positive': positive_annotations,
'negative': negative_annotations
}
self.jpeg_images = jpeg_images
self.annotation_dict = annotation_dict
def __init__(self, root_dir, transform=None):
super(BBoxRegressionDataset, self).__init__()
self.transform = transform
samples = util.parse_car_csv(root_dir)
jpeg_list = list()
# 保存{'image_id': ?, 'positive': ?, 'bndbox': ?}
box_list = list()
for i in range(len(samples)):
sample_name = samples[i]
jpeg_path = os.path.join(root_dir, 'JPEGImages',
sample_name + '.jpg')
bndbox_path = os.path.join(root_dir, 'bndboxs',
sample_name + '.csv')
positive_path = os.path.join(root_dir, 'positive',
sample_name + '.csv')
jpeg_list.append(cv2.imread(jpeg_path))
bndboxes = np.loadtxt(bndbox_path, dtype=np.int, delimiter=' ')
positives = np.loadtxt(positive_path, dtype=np.int, delimiter=' ')
if len(positives.shape) == 1:
bndbox = self.get_bndbox(bndboxes, positives)
box_list.append({
'image_id': i,
'positive': positives,
'bndbox': bndbox
})
else:
for positive in positives:
bndbox = self.get_bndbox(bndboxes, positive)
box_list.append({
'image_id': i,
'positive': positive,
'bndbox': bndbox
})
self.jpeg_list = jpeg_list
self.box_list = box_list
for name in ['train', 'val']:
src_root_dir = os.path.join(car_root_dir, name)
src_annotation_dir = os.path.join(src_root_dir, 'Annotations')
src_jpeg_dir = os.path.join(src_root_dir, 'JPEGImages')
dst_root_dir = os.path.join(finetune_root_dir, name)
dst_annotation_dir = os.path.join(dst_root_dir, 'Annotations')
dst_jpeg_dir = os.path.join(dst_root_dir, 'JPEGImages')
check_dir(dst_root_dir)
check_dir(dst_annotation_dir)
check_dir(dst_jpeg_dir)
total_num_positive = 0
total_num_negative = 0
samples = parse_car_csv(src_root_dir)
# 复制csv文件
src_csv_path = os.path.join(src_root_dir, 'car.csv')
dst_csv_path = os.path.join(dst_root_dir, 'car.csv')
shutil.copyfile(src_csv_path, dst_csv_path)
for sample_name in samples:
since = time.time()
src_annotation_path = os.path.join(src_annotation_dir,
sample_name + '.xml')
src_jpeg_path = os.path.join(src_jpeg_dir, sample_name + '.jpg')
# 获取正负样本
positive_list, negative_list = parse_annotation_jpeg(
src_annotation_path, src_jpeg_path, gs)
total_num_positive += len(positive_list)
total_num_negative += len(negative_list)
def __init__(self, root_dir, transform=None):
samples = parse_car_csv(root_dir)
jpeg_images = list()
positive_list = list()
negative_list = list()
for idx in range(len(samples)):
sample_name = samples[idx]
jpeg_images.append(
cv2.imread(
os.path.join(root_dir, 'JPEGImages',
sample_name + ".jpg")))
positive_annotation_path = os.path.join(root_dir, 'Annotations',
sample_name + '_1.csv')
positive_annotations = np.loadtxt(positive_annotation_path,
dtype=np.int,
delimiter=' ')
# 考虑csv文件为空或者仅包含单个标注框
if len(positive_annotations.shape) == 1:
# 单个标注框坐标
if positive_annotations.shape[0] == 4:
positive_dict = dict()
positive_dict['rect'] = positive_annotations
positive_dict['image_id'] = idx
# positive_dict['image_name'] = sample_name
positive_list.append(positive_dict)
else:
for positive_annotation in positive_annotations:
positive_dict = dict()
positive_dict['rect'] = positive_annotation
positive_dict['image_id'] = idx
# positive_dict['image_name'] = sample_name
positive_list.append(positive_dict)
negative_annotation_path = os.path.join(root_dir, 'Annotations',
sample_name + '_0.csv')
negative_annotations = np.loadtxt(negative_annotation_path,
dtype=np.int,
delimiter=' ')
# 考虑csv文件为空或者仅包含单个标注框
if len(negative_annotations.shape) == 1:
# 单个标注框坐标
if negative_annotations.shape[0] == 4:
negative_dict = dict()
negative_dict['rect'] = negative_annotations
negative_dict['image_id'] = idx
# negative_dict['image_name'] = sample_name
negative_list.append(negative_dict)
else:
for negative_annotation in negative_annotations:
negative_dict = dict()
negative_dict['rect'] = negative_annotation
negative_dict['image_id'] = idx
# negative_dict['image_name'] = sample_name
negative_list.append(negative_dict)
self.transform = transform
self.jpeg_images = jpeg_images
self.positive_list = positive_list
self.negative_list = negative_list
classifier_car_train_dir = '../../data/finetune_car/train'
# positive
positive_annotation_dir = os.path.join(classifier_car_train_dir, 'Annotations')
dst_root_dir = '../../data/bbox_regression/'
dst_jpeg_dir = os.path.join(dst_root_dir, 'JPEGImages')
dst_bndbox_dir = os.path.join(dst_root_dir, 'bndboxs')
dst_positive_dir = os.path.join(dst_root_dir, 'positive')
util.check_dir(dst_root_dir)
util.check_dir(dst_jpeg_dir)
util.check_dir(dst_bndbox_dir)
util.check_dir(dst_positive_dir)
samples = util.parse_car_csv(voc_car_train_dir)
res_samples = list()
total_positive_num = 0
for sample_name in samples:
# 提取正样本边界框坐标（IoU>=0.5）
positive_annotation_path = os.path.join(positive_annotation_dir, sample_name + '_1.csv')
positive_bndboxes = np.loadtxt(positive_annotation_path, dtype=np.int, delimiter=' ')
# 提取标注边界框
gt_annotation_path = os.path.join(gt_annotation_dir, sample_name + '.xml')
bndboxs = util.parse_xml(gt_annotation_path)
# 计算符合条件（IoU>0.6）的候选建议
positive_list = list()
if len(positive_bndboxes.shape) == 1 and len(positive_bndboxes) != 0:
scores = util.iou(positive_bndboxes, bndboxs)
if np.max(scores) > 0.6:
positive_list.append(positive_bndboxes)
