"""
使用VGG16模型测试图片识别
20221116 byp
程序来源：https://blog.csdn.net/java_raylu/article/details/123536858
注意：keras提示有红色波浪线，但是程序依然可以执行
增加前3种预测结果
"""
from tensorflow.keras import applications
from tensorflow.keras import applications
# decode_predictions 对预测结果进行处理，解码
from tensorflow.keras.applications.vgg16 import decode_predictions
# preprocess_input处理输入图片
from tensorflow.keras.applications.vgg16 import preprocess_input
from tensorflow import keras

from tensorflow.keras.preprocessing.image import img_to_array
from tensorflow.keras.preprocessing.image import load_img


# VGG 图片输入大小要求 224*224*3
# VGG 有1.4亿个参数

# 获取已训练好的模型
def predict(my_model, images):
    y_lables = []
    for image in images:
        y_predictions = my_model.predict(image)
        # print(y_predictions)
        y_lable = decode_predictions(y_predictions)
        y_lables.append(y_lable)

    return y_lables


# 图片读取处理
def load_img1(files_path):
    images = []
    # 加载图片
    for path in files_path:
        image = load_img(path, target_size=(224, 224))
        # 图片进行数组转换
        image = img_to_array(image)
        print('-----')
        print(image.shape)
        # 形状修改,输入到卷积需要四维[1,224,224,3],1代表batch,如果多张图片就使用多张
        image = image.reshape(1, image.shape[0], image.shape[1], image.shape[2])
        print(image.shape)
        # print(image)
        # 对图片进行归一化处理
        images.append(preprocess_input(image))
    return images


# 输入到模型进行预测

if __name__ == '__main__':
    import os

    # 获取当前路径
    path = os.getcwd()
    files = os.listdir(os.path.join(path, "tiger"))
    file_count = len(files)
    print(file_count)
    files_path = []
    for i in range(file_count):
        print(i)
        files_path.append(os.path.join(path, 'tiger', files[i]))

    print(files_path)
    my_model = applications.vgg16.VGG16()
    images = load_img1(files_path)
    print('123')
    y_predicts = predict(my_model, images)
    print(y_predicts)
    i = 0
    for y_predict in y_predicts:
        i = i + 1
        print('图片', i, '预测结果是:%s,%.2f' % (y_predict[0][0][1], y_predict[0][0][2] * 100), "%;",
              '%s,%.2f' % (y_predict[0][1][1], y_predict[0][1][2] * 100), "%;",
              '%s,%.2f' % (y_predict[0][2][1], y_predict[0][2][2] * 100), "%")