def convert_line_to_text_dataset(anno_filename, label_out_dir, show=True):
file_processing.create_dir(label_out_dir)
boxes_label_lists = file_processing.read_lines_image_labels(anno_filename)
for image_id, box, label in boxes_label_lists:
filename = image_id[:-len('jpg')] + "txt"
content_list = [[c] + r for c, r in zip(label, box)]
path = os.path.join(label_out_dir, filename)
file_processing.write_data(path, content_list, mode='w')
def create_embedding(model_path, emb_face_dir):
face_net = face_recognition.facenetEmbedding(model_path)
image_list, names_list = file_processing.gen_files_labels(emb_face_dir,
postfix='jpg')
images = image_processing.get_images(image_list, 160, 160, whiten=True)
compare_emb = face_net.get_embedding(images)
h5file = h5py.File('face.h5', 'w')
h5file['X_train'] = compare_emb
h5file.close()
file_processing.write_data('name.txt', image_list, model='w')
def convert_facebody_to_textdataset(image_list,
annotations_dir,
label_out_dir,
class_names,
coordinatesType,
show=False):
'''
label data format：
SSD = [label_id,x,y,w,h]
YOLO = [label_id,x_center/img_width ,y_center/img_height ,width/img_width ,height/img_height]
MMDET= [img_width,img_height,label_id,x,y,w,h]
:param image_list: 图片列表
:param annotations_dir: 图片对应annotations所在目录
:param label_out_dir: label输出目录
:param class_names:
:param coordinatesType: 坐标类型：SSD,YOLO,MMDET格式
:param show: 显示
:return:
'''
if not os.path.exists(label_out_dir):
os.makedirs(label_out_dir)
name_id_list = []
nums = len(image_list)
for i, image_path in enumerate(image_list):
name_id = os.path.basename(image_path)[:-len(".jpg")]
ann_name = name_id + '.json'
annotations_file = os.path.join(annotations_dir, ann_name)
if not os.path.exists(image_path):
print("no image_dict:{}".format(image_path))
continue
if not os.path.exists(annotations_file):
print("no annotations:{}".format(annotations_file))
continue
out_file = os.path.join(label_out_dir, name_id + ".txt")
# rects, class_name, class_id = pascal_voc.get_annotation(annotations_file, class_names, coordinatesType)
image = image_processing.read_image(image_path)
image_shape = image.shape
rects, class_name, class_id = face_body.get_annotation(
annotations_file, class_names, image_shape, coordinatesType)
if len(rects) == 0 or len(class_name) == 0 or len(class_id) == 0:
print("no class in annotations:{}".format(annotations_file))
continue
content_list = [[c] + r for c, r in zip(class_id, rects)]
name_id_list.append(name_id)
file_processing.write_data(out_file, content_list, mode='w')
if show:
image = image_processing.read_image(image_path)
image_processing.show_image_rects_text("image_dict", image, rects,
class_name)
if i % 10 == 0 or i == len(image_list) - 1:
print("processing image_dict:{}/{}".format(i, len(image_list) - 1))
return name_id_list
def save_label_set(filename, label_name_set):
'''
保存label数据
:param filename:
:param label_name_set:
:return:
'''
# 将字符集char_set转为整形的标签集合
# label_set=list(range(0,len(label_name_set)))
content_list = []
for label_index, name in enumerate(label_name_set):
# content=name +" "+str(label_index)
content = name
content_list.append(content)
file_processing.write_data(filename, content_list, model='w')
def create_embedding(model_path, emb_face_dir, out_emb_path, out_filename):
'''
产生embedding数据库，这些embedding其实就是人脸特征
'''
print('#3rd--FaceNet人脸库')
face_net = face_recognition.facenetEmbedding(model_path)
image_list,names_list=file_processing.gen_files_labels(emb_face_dir,postfix='jpg')
images= image_processing.get_images(image_list,resize_height,resize_width,whiten=True)
compare_emb = face_net.get_embedding(images)
np.save(out_emb_path, compare_emb)
file_processing.write_data(out_filename, names_list, model='w')
def convert_annotation_list(annotations_list,
image_dir,
label_out_dir,
class_names,
image_type='.jpg',
show=True):
'''
:param annotations_list:annotations列表
:param image_dir:图片所在路径
:param label_out_dir:输出label目录
:param class_names:
:param image_type:图片的类型，如.jpg ,.png
:param show:
:return:
'''
if not os.path.exists(label_out_dir):
os.makedirs(label_out_dir)
name_id_list = []
nums = len(annotations_list)
for i, annotations_file in enumerate(annotations_list):
name_id = os.path.basename(annotations_file)[:-len(".xml")]
image_name = name_id + image_type
image_path = os.path.join(image_dir, image_name)
if not os.path.exists(image_path):
print("no image:{}".format(image_path))
continue
if not os.path.exists(annotations_file):
print("no annotations:{}".format(annotations_file))
continue
out_file = os.path.join(label_out_dir, name_id + ".txt")
rects, class_name, class_id = pascal_voc.get_annotation(
annotations_file, class_names)
content_list = [[c] + r for c, r in zip(class_id, rects)]
name_id_list.append(name_id)
file_processing.write_data(out_file, content_list, mode='w')
if show:
image = image_processing.read_image(image_path)
image_processing.show_image_rects_text("image", image, rects,
class_name)
if i % 100 == 0 or i == nums - 1:
print("processing {}/{}".format(i + 1, nums))
return name_id_list
def create_face_embedding(model_path, dataset_path, out_emb_path,
out_filename):
'''
:param model_path: faceNet模型路径
:param dataset_path: 人脸数据库路径，每一类单独一个文件夹
:param out_emb_path: 输出embeddings的路径
:param out_filename: 输出与embeddings一一对应的标签
:return: None
'''
files_list, names_list = file_processing.gen_files_labels(dataset_path,
postfix='jpg')
embeddings, label_list = get_face_embedding(model_path, files_list,
names_list)
print("label_list:{}".format(label_list))
print("have {} label".format(len(label_list)))
embeddings = np.asarray(embeddings)
np.save(out_emb_path, embeddings)
file_processing.write_data(out_filename, label_list, model='w')
def create_dataset(out_dir, nums, filename, char_set, captcha_height,
captcha_width, captcha_size):
'''
产生样本
:param out_dir: 数据集图片保存目录
:param nums: 产生数据样本个数
:param filename: 保存数据txt文件
:param char_set: 字符数据集
:param captcha_height: 验证码height
:param captcha_width: 验证码width
:param captcha_size: 验证码大小
:return:None
'''
if not os.path.exists(out_dir):
os.mkdir(out_dir)
# 产生一个验证码样本并显示
i = 0
while i < nums:
text, image = gen_captcha_text_and_image(char_set=char_set,
captcha_height=captcha_height,
captcha_width=captcha_width,
captcha_size=captcha_size)
# 产生的验证码图并不一定等于image_height*image_width
if image.shape != (image_height, image_width, 3):
continue
if i == 0:
image_processing.cv_show_image(text, image) # 显示验证码
image_name = str(i) + "_" + text + ".jpg"
image_path = out_dir + "/" + image_name
print(image_path)
image_processing.save_image(image_path, image, toUINT8=False)
text = [c for c in text]
label_list = file_processing.label_encode(text, char_set)
content = [image_name] + label_list
content = ' '.join('%s' % id for id in content)
file_processing.write_data(filename, [content], model='a')
i += 1
def create_face_embedding_for_bzl(model_path, dataset_path, out_emb_path,
out_filename):
'''
:param model_path: faceNet模型路径
:param dataset_path: 人脸数据库路径，图片命名方式：张三_XXX_XXX.jpg,其中“张三”即为label
:param out_emb_path: 输出embeddings的路径
:param out_filename: 输出与embeddings一一对应的标签
:return: None
'''
image_list = file_processing.get_images_list(dataset_path,
postfix=['*.jpg', '*.png'])
names_list = []
for image_path in image_list:
basename = os.path.basename(image_path)
names = basename.split('_')[0]
names_list.append(names)
embeddings, label_list = get_face_embedding(model_path, image_list,
names_list)
print("label_list:{}".format(label_list))
print("have {} label".format(len(label_list)))
embeddings = np.asarray(embeddings)
np.save(out_emb_path, embeddings)
file_processing.write_data(out_filename, label_list, mode='w')
def create_embedding(model_path, emb_face_dir, out_emb_path, out_filename):
'''
产生embedding数据库，保存在out_data_path中，这些embedding其实就是人脸的特征
:param model_path:
:param emb_face_dir:
:param out_emb_path:
:param out_filename:
:return:
'''
face_net = face_recognition.facenetEmbedding(model_path)
# image_list=file_processing.get_files_list(emb_face_dir,postfix='jpg')
image_list, names_list = file_processing.gen_files_labels(emb_face_dir,
postfix='jpg')
images = image_processing.get_images(image_list,
resize_height,
resize_width,
whiten=True)
compare_emb = face_net.get_embedding(images)
np.save(out_emb_path, compare_emb)
# 可以选择保存image_list或者names_list作为人脸的标签
# 测试时建议保存image_list，这样方便知道被检测人脸与哪一张图片相似
file_processing.write_data(out_filename, image_list, model='w')
def save_pair_data(filename, content_list):
file_processing.write_data(filename, content_list, mode='w')
else:
if select_nums > num_pair_issame_1:
raise Exception(
"pair_nums({}) must be less than num_pair_issame_1({})".format(
select_nums, num_pair_issame_1))
np.random.seed(100)
index_0 = np.random.permutation(num_pair_issame_0)[:select_nums] # 打乱后的行号
index_1 = np.random.permutation(num_pair_issame_1)[:select_nums] # 打乱后的行号
pair_issame_0 = pair_issame_0[index_0, :] # 获取打乱后的训练数据
pair_issame_1 = pair_issame_1[index_1, :] # 获取打乱后的训练数据
pair_issame = np.concatenate([pair_issame_0, pair_issame_1], axis=0)
print("pair_issame_0 nums:{}".format(len(pair_issame_0)))
print("pair_issame_1 nums:{}".format(len(pair_issame_1)))
# image_list1 = pair_issame[:, 0]
# image_list2 = pair_issame[:, 1]
# issame_list = pair_issame[:, 2]
print("have {} pairs".format(len(pair_issame)))
return pair_issame
if __name__ == "__main__":
# NVR VAL faceDataset
dataset = '/media/dm/dm2/XMC/FaceDataset/X4/X4_Face20_Crop/'
# dataset = '/media/dm/dm1/FaceDataset/X4/DMAI_Alig/'
# lexue
image_dir = dataset + "trainval"
pair_filename = dataset + "x4_pair_data.txt"
pair_issame = create_pair_data(image_dir, pair_num=0)
file_processing.write_data(pair_filename, pair_issame, mode='w')
def convert_voc_to_textdataset_for_annotation(annotations_list,
image_dir,
label_out_dir,
class_names,
coordinatesType,
image_type='.jpg',
labelType="class_id",
show=True):
'''
coordinatesType:
SSD = [label_id,x,y,w,h]
xywh = [label_id,x,y,w,h]
xyxy = [label_id,xmin,ymin,xmax,ymax]
xxyy = [label_id,xmin,xmax,ymin,ymax]
YOLO = [label_id,x_center/img_width ,y_center/img_height ,width/img_width ,height/img_height]
MMDET= [img_width,img_height,label_id,x,y,w,h]
:param annotations_list:annotations列表
:param image_dir:图片所在路径
:param label_out_dir:输出label目录
:param class_names:
:param image_type:图片的类型，如.jpg ,.png
:param labelType:class_name,class_id
:param show:
:return:
'''
if not os.path.exists(label_out_dir):
os.makedirs(label_out_dir)
name_id_list = []
nums = len(annotations_list)
for i, annotations_file in enumerate(annotations_list):
name_id = os.path.basename(annotations_file)[:-len(".xml")]
image_name = name_id + image_type
image_path = os.path.join(image_dir, image_name)
if not os.path.exists(image_path):
print("no image_dict:{}".format(image_path))
continue
if not os.path.exists(annotations_file):
print("no annotations:{}".format(annotations_file))
continue
out_file = os.path.join(label_out_dir, name_id + ".txt")
rects, class_name, class_id = pascal_voc.get_annotation(
annotations_file, class_names, coordinatesType=coordinatesType)
if not rects:
continue
if labelType == "class_name":
label = class_name
elif labelType == "class_id":
label = class_id
content_list = [[c] + r for c, r in zip(label, rects)]
name_id_list.append(name_id)
file_processing.write_data(out_file, content_list, mode='w')
if show:
image = image_processing.read_image(image_path)
# rect_image = image_processing.get_rects_image(image_dict,rects)
# save_root=DATASET_ROOT+"/trainval_faces"
# image_processing.save_image_lable_dir(save_root, rect_image, class_name,i)
image_processing.show_image_rects_text("image_dict", image, rects,
class_name)
if i % 10 == 0 or i == len(annotations_list) - 1:
print("processing image_dict:{}/{}".format(
i,
len(annotations_list) - 1))
return name_id_list
