def convert_image_to_bcolz(pair_filename,
image_dir,
save_dir,
input_size=[112, 112]):
from torchvision import transforms as trans
import bcolz
transform = trans.Compose(
[trans.ToTensor(),
trans.Normalize([0.5, 0.5, 0.5], [0.5, 0.5, 0.5])])
if not os.path.exists(save_dir):
os.makedirs(save_dir)
faces_list1, faces_list2, issames_data = read_pair_data(pair_filename)
print("have {} pair".format(len(issames_data)))
print("have {} pair".format(len(faces_list1)))
issames_data = np.array(issames_data)
issames_data = np.where(issames_data > 0, True, False)
data = bcolz.fill(shape=[
len(faces_list1) + len(faces_list2), 3, input_size[0], input_size[1]
],
dtype=np.float32,
rootdir=save_dir,
mode='w')
for i, (face1_path, face2_path,
issame) in enumerate(zip(faces_list1, faces_list2, issames_data)):
# pred_id, pred_scores = faceRec.predict(faces)
# 或者使用get_faces_embedding()获得embedding，再比较compare_embedding()
if image_dir:
face1_path = os.path.join(image_dir, face1_path)
face2_path = os.path.join(image_dir, face2_path)
face1 = image_processing.read_image_gbk(face1_path, colorSpace="BGR")
face2 = image_processing.read_image_gbk(face2_path, colorSpace="BGR")
face1 = image_processing.resize_image(face1,
resize_height=input_size[0],
resize_width=input_size[1])
face2 = image_processing.resize_image(face2,
resize_height=input_size[0],
resize_width=input_size[1])
# img = cv2.cvtColor(img, cv2.COLOR_RGB2BGR)
# image_processing.cv_show_image("image_dict",img)
face1 = Image.fromarray(face1.astype(np.uint8))
face2 = Image.fromarray(face2.astype(np.uint8))
data[i * 2, ...] = transform(face1)
data[i * 2 + 1, ...] = transform(face2)
if i % 100 == 0:
print('loading bin', i)
print(data.shape)
np.save(str(save_dir) + '_list', issames_data)
def face_recognition_image(model_path, dataset_path, filename, image_path):
# 加载数据库的数据
dataset_emb, names_list = load_dataset(dataset_path, filename)
# 初始化mtcnn人脸检测
face_detect = face_recognition.Facedetection()
# 初始化facenet
face_net = face_recognition.facenetEmbedding(model_path)
image = image_processing.read_image_gbk(image_path)
# 获取 判断标识 bounding_box crop_image
bboxes, landmarks = face_detect.detect_face(image)
bboxes, landmarks = face_detect.get_square_bboxes(bboxes,
landmarks,
fixed="height")
if bboxes == [] or landmarks == []:
print("-----no face")
exit(0)
# print("-----image have {} faces".format(len(bboxes)))
face_images = image_processing.get_bboxes_image(image, bboxes,
resize_height,
resize_width)
face_images = image_processing.get_prewhiten_images(face_images)
pred_emb = face_net.get_embedding(face_images)
pred_name, pred_score = compare_embadding(pred_emb, dataset_emb,
names_list)
# 在图像上绘制人脸边框和识别的结果
show_info = [n + ':' + str(s)[:5] for n, s in zip(pred_name, pred_score)]
print(show_info)
def face_recognition_image(model_path, dataset_path, filename, image_path):
# 加载数据库的数据
dataset_emb, names_list = load_dataset(dataset_path, filename)
# 初始化mtcnn人脸检测
face_detect = face_recognition.FaceDetection()
# 初始化facenet
face_net = face_recognition.facenetEmbedding(model_path)
# 读取待检图片
image = image_processing.read_image_gbk(image_path)
print("image_processing.read_image_gbk:", type(image),
image.shape) # <class 'numpy.ndarray'>, (616, 922, 3)，（高，宽，通道）
# 获取 判断标识 bounding_box crop_image
bboxes, landmarks = face_detect.detect_face(image)
bboxes, landmarks = face_detect.get_square_bboxes(
bboxes, landmarks, fixed="height") # 以高为基准，获得等宽的举行
if bboxes == [] or landmarks == []:
print("-----no face")
exit(0)
print("-----image have {} faces".format(len(bboxes)))
face_images = image_processing.get_bboxes_image(
image, bboxes, resize_height, resize_width) # 按照bboxes截取矩形框
face_images = image_processing.get_prewhiten_images(face_images) # 图像归一化
pred_emb = face_net.get_embedding(face_images) # 获取facenet特征
pred_name, pred_score = compare_embadding(pred_emb, dataset_emb,
names_list)
# 在图像上绘制人脸边框和识别的结果
show_info = [n + ':' + str(s)[:5] for n, s in zip(pred_name, pred_score)]
image_processing.show_image_bboxes_text("face_reco", image, bboxes,
show_info)
def get_face_embedding(model_path,files_list, names_list):
# 获得embedding数据
colorSpace="RGB"
face_detect = face_rec.FaceDetection()
face_net = face_rec.FacenetEmbedding(model_path)
embeddings=[]
label_list=[]
for image_path, name in zip(files_list, names_list):
print("processing image :{}".format(image_path))
image = image_processing.read_image_gbk(image_path, colorSpace=colorSpace)
if not isinstance(image, np.ndarray): continue
bboxes, landmarks = face_detect.detect_face(image)
bboxes, landmarks =face_detect.get_square_bboxes(bboxes, landmarks,fixed="height")
if bboxes == [] or landmarks == []:
print("-----no face")
continue
if len(bboxes) >= 2 or len(landmarks) >= 2:
print("-----image have {} faces".format(len(bboxes)))
continue
face_images = image_processing.get_bboxes_image(image, bboxes, resize_height, resize_width)
face_images = image_processing.get_prewhiten_images(face_images,normalization=True)
pred_emb = face_net.get_embedding(face_images)
embeddings.append(pred_emb)
label_list.append(name)
return embeddings,label_list
def label_test(image_dir, filename, class_names):
basename = os.path.basename(filename)[:-len('.txt')] + ".bmp"
image_path = os.path.join(image_dir, basename)
image = image_processing.read_image_gbk(image_path)
data = file_processing.read_data(filename, split=" ")
label_list, rect_list = file_processing.split_list(data, split_index=1)
label_list = [l[0] for l in label_list]
name_list = file_processing.decode_label(label_list, class_names)
image_processing.show_image_rects_text("object2", image, rect_list,
name_list)
def get_face_embedding(model_path, files_list, names_list):
'''
获得embedding数据
:param files_list: 图像列表
:param names_list: 与files_list一一的名称列表
:return:
'''
# 转换颜色空间RGB or BGR
colorSpace = "RGB"
# 初始化mtcnn人脸检测
face_detect = face_recognition.FaceDetection()
# 初始化facenet
face_net = face_recognition.facenetEmbedding(model_path)
embeddings = [] # 用于保存人脸特征数据库
label_list = [] # 保存人脸label的名称，与embeddings一一对应
for image_path, name in zip(files_list, names_list):
print("processing image: {}".format(image_path))
image = image_processing.read_image_gbk(image_path,
colorSpace=colorSpace)
# 进行人脸检测，获得bounding_box
bboxes, landmarks = face_detect.detect_face(image)
bboxes, landmarks = face_detect.get_square_bboxes(bboxes,
landmarks,
fixed="height")
# image_processing.show_image_boxes("image",image,bboxes)
if bboxes == [] or landmarks == []:
print("-----no face")
continue
if len(bboxes) >= 2 or len(landmarks) >= 2:
print("-----image total {} faces".format(len(bboxes)))
continue
# 获得人脸区域
face_images = image_processing.get_bboxes_image(
image, bboxes, resize_height, resize_width)
# 人脸预处理，归一化
face_images = image_processing.get_prewhiten_images(face_images,
normalization=True)
# 获得人脸特征
pred_emb = face_net.get_embedding(face_images)
embeddings.append(pred_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
# label_list.append(image_path)
label_list.append(name)
return embeddings, label_list
def select_image(self):
img_name, img_type = QFileDialog.getOpenFileName(
self, "打开图片", "./data/test", "*.jpg;;*.png;;All Files(*)")
if not img_name: return
fixed_img = osjoin(cache_path, img_name.split('/')[-1])
shutil.copy(img_name, fixed_img)
image = image_processing.read_image_gbk(fixed_img)
image = image_fix(image)
cv2.imwrite(fixed_img, image)
iw, ih = image.shape[1], image.shape[0]
rw, rh = self.rec_result.width(), self.rec_result.height()
w, h = image_processing.scaled_to(iw, ih, rw, rh)
img = QtGui.QPixmap(fixed_img).scaled(w, h)
self.rec_result.setAlignment(Qt.AlignCenter)
self.rec_result.setPixmap(img)
os.remove(fixed_img)
def convert_image_format(image_dir,
dest_dir,
resize_width=None,
dest_format='.jpg'):
image_id = file_processing.get_sub_directory_list(image_dir)
for id in image_id:
image_list = file_processing.get_files_list(
os.path.join(image_dir, id),
postfix=['*.jpg', "*.jpeg", '*.png', "*.JPG"])
print("processing :{}".format(id))
for src_path in image_list:
basename = os.path.basename(src_path).split('.')[0]
image = image_processing.read_image_gbk(src_path,
resize_width=resize_width)
dest_path = file_processing.create_dir(dest_dir, id,
basename + dest_format)
file_processing.create_file_path(dest_path)
image_processing.save_image(dest_path, image)
def classify_faces(dataset_path):
# 将人脸图像分为0，1和n>1张脸三类，存到三个文件夹下，供后续人工简单筛选用
face_detect = face_rec.FaceDetection()
classify_root_path = './data/classify'
classify_path = [osjoin(classify_root_path, x) for x in ('NA', '0', '1', 'n')]
for tp in classify_path:
if os.path.exists(tp): continue
os.mkdir(tp)
paths = os.listdir(dataset_path)
for img_path in paths:
real_path = osjoin(dataset_path, img_path)
image = image_processing.read_image_gbk(real_path, colorSpace='RGB')
if not isinstance(image, np.ndarray):
out_put_path = osjoin(classify_path[0], img_path)
else:
bboxes, landmarks = face_detect.detect_face(image)
bboxes, landmarks = face_detect.get_square_bboxes(bboxes, landmarks,fixed="height")
if bboxes == [] or landmarks == []:
out_put_path = osjoin(classify_path[1], img_path)
elif len(bboxes) >= 2 or len(landmarks) >= 2:
out_put_path = osjoin(classify_path[3], img_path)
else:
out_put_path = osjoin(classify_path[2], img_path)
shutil.copy(real_path, out_put_path)
def face_recognition_for_bzl(model_path, test_dataset, filename):
# 加载数据库的数据
dataset_emb, names_list = predict.load_dataset(dataset_path, filename)
print("loadind dataset...\n names_list:{}".format(names_list))
# 初始化mtcnn人脸检测
face_detect = face_recognition.Facedetection()
# 初始化facenet
face_net = face_recognition.facenetEmbedding(model_path)
#获得测试图片的路径和label
filePath_list, label_list = file_processing.gen_files_labels(test_dataset)
label_list = [name.split('_')[0] for name in label_list]
print("filePath_list:{},label_list{}".format(len(filePath_list),
len(label_list)))
right_num = 0
wrong_num = 0
detection_num = 0
test_num = len(filePath_list)
for image_path, label_name in zip(filePath_list, label_list):
print("image_path:{}".format(image_path))
# 读取图片
image = image_processing.read_image_gbk(image_path)
# 人脸检测
bboxes, landmarks = face_detect.detect_face(image)
bboxes, landmarks = face_detect.get_square_bboxes(bboxes,
landmarks,
fixed="height")
if bboxes == [] or landmarks == []:
print("-----no face")
continue
if len(bboxes) >= 2 or len(landmarks) >= 2:
print("-----image have {} faces".format(len(bboxes)))
continue
# 获得人脸框区域
face_images = image_processing.get_bboxes_image(
image, bboxes, resize_height, resize_width)
face_images = image_processing.get_prewhiten_images(face_images,
normalization=True)
# face_images = image_processing.get_prewhiten_images(face_images,normalization=True)
pred_emb = face_net.get_embedding(face_images)
pred_name, pred_score = predict.compare_embadding(pred_emb,
dataset_emb,
names_list,
threshold=1.3)
# 在图像上绘制人脸边框和识别的结果
# show_info = [n + ':' + str(s)[:5] for n, s in zip(pred_name, pred_score)]
# image_processing.show_image_text("face_recognition", image, bboxes, show_info)
index = 0
pred_name = pred_name[index]
pred_score = pred_score[index]
if pred_name == label_name:
right_num += 1
else:
wrong_num += 1
detection_num += 1
print(
"-------------label_name:{},pred_name:{},score:{:3.4f},status:{}".
format(label_name, pred_name, pred_score,
(label_name == pred_name)))
# 准确率
accuracy = right_num / detection_num
# 漏检率
misdetection = (test_num - detection_num) / test_num
print("-------------right_num/detection_num:{}/{},accuracy rate:{}".format(
right_num, detection_num, accuracy))
print(
"-------------misdetection/all_num:{}/{},misdetection rate:{}".format(
(test_num - detection_num), test_num, misdetection))