def face_predict(self, image):
#依然是根据后端系统确定维度顺序
if K.image_dim_ordering() == 'th' and image.shape != (1, 1, IMAGE_SIZE,
IMAGE_SIZE):
image = resize_image(image) #尺寸必须与训练集一致都应该是IMAGE_SIZE x IMAGE_SIZE
image = image.reshape(
(1, 1, IMAGE_SIZE, IMAGE_SIZE)) #与模型训练不同，这次只是针对1张图片进行预测
elif K.image_dim_ordering() == 'tf' and image.shape != (1, IMAGE_SIZE,
IMAGE_SIZE, 1):
image = resize_image(image)
image = image.reshape((1, IMAGE_SIZE, IMAGE_SIZE, 1))
#浮点并归一化
image = image.astype('float32')
image /= 255
#给出输入属于各个类别的概率，我们是二值类别，则该函数会给出输入图像属于0和1的概率各为多少
result = self.model.predict_proba(image)
print('result:', result)
#给出类别预测：0或者1
result = self.model.predict_classes(image)
#返回类别预测结果
return result[0]
def face_predict(self, image):
#依然是根據後端系統確定維度順序
if K.common.image_dim_ordering() == 'th' and image.shape != (
1, 3, IMAGE_SIZE, IMAGE_SIZE):
image = resize_image(image) #尺寸必須與訓練集一致都應該是IMAGE_SIZE x IMAGE_SIZE
image = image.reshape(
(1, 3, IMAGE_SIZE, IMAGE_SIZE)) #與模型訓練不同，這次只是針對1張圖片進行預測
elif K.common.image_dim_ordering() == 'tf' and image.shape != (
1, IMAGE_SIZE, IMAGE_SIZE, 3):
image = resize_image(image)
image = image.reshape((1, IMAGE_SIZE, IMAGE_SIZE, 3))
#浮點並歸一化
image = image.astype('float32')
image /= 255
#給出輸入屬於各個類別的概率，我們是二值類別，則該函式會給出輸入影象屬於0和1的概率各為多少
result = self.model.predict_proba(image)
print('result:', result)
#給出類別預測：0或者1
result = self.model.predict_classes(image)
#返回類別預測結果
return result[0]
def face_predict(self, image):
# K.set_image_dim_ordering('tf') --> K.set_image_data_format('channels_last')
# K.set_image_dim_ordering('th') --> K.set_image_data_format('channels_first')
# K.image_dim_ordering() == 'tf' --> K.image_data_format() == 'channels_last'
# K.image_dim_ordering() == 'th' --> K.image_data_format() == 'channels_first'
# 依然是根据后端系统确定维度顺序
if K.image_data_format() == 'channels_first' and image.shape != (
1, 3, IMAGE_SIZE, IMAGE_SIZE):
image = resize_image(
image) # 尺寸必须与训练集一致都应该是IMAGE_SIZE x IMAGE_SIZE
image = image.reshape(
(1, 3, IMAGE_SIZE, IMAGE_SIZE)) # 与模型训练不同，这次只是针对1张图片进行预测
elif K.image_data_format() == 'channels_last' and image.shape != (
1, IMAGE_SIZE, IMAGE_SIZE, 3):
image = resize_image(image)
image = image.reshape((1, IMAGE_SIZE, IMAGE_SIZE, 3))
# 浮点并归一化
image = image.astype('float32')
image /= 255
# 给出输入属于各个类别的概率
result_probability = self.model.predict_proba(image)
# print('result:', result_probability)
# 给出类别预测(改）
if max(result_probability[0]) >= 0.9:
result = self.model.predict_classes(image)
print('result:', result)
# 返回类别预测结果
return result[0]
else:
return -1
def face_predict(self, image):
# 依然是根据后端系统确定维度顺序
if K.image_data_format() == 'channels_first' and image.shape != (1, 3, IMAGE_SIZE, IMAGE_SIZE):
image = resize_image(image) # 尺寸必须与训练集一致都应该是IMAGE_SIZE x IMAGE_SIZE
image = image.reshape((1, 3, IMAGE_SIZE, IMAGE_SIZE)) # 与模型训练不同，这次只是针对1张图片进行预测
elif K.image_data_format() == 'channels_last' and image.shape != (1, IMAGE_SIZE, IMAGE_SIZE, 3):
image = resize_image(image)
image = image.reshape((1, IMAGE_SIZE, IMAGE_SIZE, 3))
# 浮点并归一化
# image = image.astype('float32')
# image /= 255
image = tf.cast(image, tf.float32) * (1. / 255) - 0.5 # 归一化
# 给出输入属于各个类别的概率
result_probability = self.model.predict_proba(image)
print('result:', result_probability)
# 给出类别预测(改）
print(max(result_probability[0]))
if max(result_probability[0]) >= 0.9:
result = self.model.predict_classes(image)
print('result:', result)
# 返回类别预测结果
return result[0]
else:
return -1
def face_predict(self, image):
# 依然是根据后端系统确定维度顺序
if K.image_dim_ordering() == 'th' and image.shape != (1, 3, IMAGE_SIZE,
IMAGE_SIZE):
image = resize_image(
image) # 尺寸必须与训练集一致都应该是IMAGE_SIZE x IMAGE_SIZE
image = image.reshape(
(1, 3, IMAGE_SIZE, IMAGE_SIZE)) # 与模型训练不同，这次只是针对1张图片进行预测
elif K.image_dim_ordering() == 'tf' and image.shape != (1, IMAGE_SIZE,
IMAGE_SIZE, 3):
image = resize_image(image)
image = image.reshape((1, IMAGE_SIZE, IMAGE_SIZE, 3))
# 浮点并归一化
image = image.astype('float32') #转换类型
image /= 255
#给出输入属于各个类别的概率
result_probability = self.model.predict_proba(image) #返回数组
#print('result:', result_probability)
#给出类别预测(改）
if max(result_probability[0]) >= 0.9: #0.9为识别相似度
result = self.model.predict_classes(image) #返回索引
print('result:', result)
#print(result_probability[0]) #输出data中每位人物的识别概率
# 返回类别预测结果
return result[0]
else:
return -1
def resize_test_image(self, image_path, image=None):
if image is None:
image = cv2.imread(image_path)
i = resize_image(image) # 需要进行预测的数据也要进行等比缩放
inp = i.reshape(
1, 64, 64,
3) # 重新确定image的形状 测试数据输入维度应当为4维 数量*IMAGESIZE*IMAGESIZE*RGB
inp = inp.astype('float32') # dtype需要一致 这里更该为 float32
return inp
else:
i = resize_image(image) # 需要进行预测的数据也要进行等比缩放
inp = i.reshape(1, 64, 64, 3) # 重新确定image的形状 测试数据输入维度应当为4维
inp = inp.astype('float32') # dtype需要一致 这里更该为 float32
return inp
def face_predict(self, image):
if K.image_dim_ordering() == 'th' and image.shape != (1, 3, IMAGE_SIZE, IMAGE_SIZE):
image = resize_image(image)
image = image.reshape((1, 3, IMAGE_SIZE, IMAGE_SIZE))
elif K.image_dim_ordering() == 'tf' and image.shape != (1, IMAGE_SIZE, IMAGE_SIZE, 3):
image = resize_image(image)
image = image.reshape((1, IMAGE_SIZE, IMAGE_SIZE, 3))
image = image.astype('float32')
image /= 255
result = self.model.predict_proba(image)
print 'result:', result
result = self.model.predict_classes(image)
return result[0]
def face_predict(self, image):
if K.image_dim_ordering() == "th" and image.shape != (1, 3, IMAGE_SIZE, IMAGE_SIZE):
image = resize_image(image)
image = image.reshape((1, 3, IMAGE_SIZE, IMAGE_SIZE))
elif K.image_dim_ordering() == "tf" and image.shape != (1, IMAGE_SIZE, IMAGE_SIZE, 3):
image = resize_image(image)
image = image.reshape((1, IMAGE_SIZE, IMAGE_SIZE, 3))
image = image.astype("float32")
image /= 255
result = self.model.predict_proba(image)
print("result: ", result)
result = self.model.predict_classes(image)
return result[0]
def face_predict(self, image):
if K.image_data_format() == 'channels_first' and image.shape != (
1, 3, IMAGE_SIZE, IMAGE_SIZE):
image = resize_image(image)
image = image.reshape((1, 3, IMAGE_SIZE, IMAGE_SIZE))
elif K.image_data_format() == 'channels_last' and image.shape != (
1, IMAGE_SIZE, IMAGE_SIZE, 3):
image = resize_image(image)
image = image.reshape((1, IMAGE_SIZE, IMAGE_SIZE, 3))
image = image.astype('float32')
image /= 255
result = self.model.predict_proba(image)
print('result:', result)
result = self.model.predict_classes(image)
return result[0]
def read_path(path_name):
for dir_item in os.listdir(path_name):
# 从初始路径开始叠加，合并成可识别的操作路径
full_path = os.path.abspath(os.path.join(path_name, dir_item))
if os.path.isdir(full_path): # 如果是文件夹，继续递归调用
read_path(full_path)
else: # 文件
if dir_item.endswith('.jpg'):
image = cv2.imread(full_path)
image = resize_image(image, IMAGE_SIZE,
IMAGE_SIZE) # 从我们的文件夹中读取图片然后进行统一规格的处理
return image
def face_predict(self, image):
face_labels = ['zhuang', 'deng', 'others', 'shang']
if image.shape != (1, IMAGE_SIZE, IMAGE_SIZE, 3):
image = resize_image(image, IMAGE_SIZE, IMAGE_SIZE)
image = image.reshape((1, IMAGE_SIZE, IMAGE_SIZE, 3))
result = self.model.predict(image)
print('result:', result[0])
result = self.model.predict_classes(image)
faceID = face_labels[result[0]]
return faceID
def face_predict_01(self, image):
image = resize_image(image)
image = image.astype('float32')
image = image / 255
result = self.model.predict(image)
return result
#-*- coding:utf-8 -*-
#author:zhangwei
import cv2
from face_dl import ModelFace
from load_dataset import resize_image
MODEL_PATH = '/home/zhangwei/face/myface.model.h5'
if __name__ == '__main__':
model = ModelFace()
model.load_mdoel()
imagepath = '/home/zhangwei/138439.jpg'
image = cv2.imread(imagepath)
image = resize_image(image)
# cv2.imshow("image" , image)
# cv2.imwrite("/home/zhangwei/1.jpg" , image)
# cv2.waitKey(10)
model = ModelFace()
model.load_mdoel()
model.face_predict_01(image)
