from keras.applications import VGG16
from keras.applications import VGG19
from keras.applications import ResNet50
from keras.applications import InceptionV3
from keras.applications import InceptionResNetV2

 image = cv2.imread(img)
 image = cv2.resize(image, self.dim)
 image = img_to_array(image)
 image = np.expand_dims(image, axis=0)

[image]
image_path=/home/fantasy/Pictures/cat.jpg
[model]
model=vgg16
[weights]
weight=imagenet

import configparser
cf = configparser.ConfigParser()
cf.read("configs.cnf")
def getOption(section, key):
  return cf.get(section, key)

# keras 提供了一些預(yù)訓(xùn)練模型，也就是開箱即用的 已經(jīng)訓(xùn)練好的模型
# 我們可以使用這些預(yù)訓(xùn)練模型來進(jìn)行圖像識(shí)別，目前的預(yù)訓(xùn)練模型大概可以識(shí)別2.2w種類型的東西
# 可用的模型：
# VGG16
# VGG19
# ResNet50
# InceptionResNetV2
# InceptionV3 
# 這些模型被集成到 keras.applications 中
# 當(dāng)我們使用了這些內(nèi)置的預(yù)訓(xùn)練模型時(shí)，模型文件會(huì)被下載到 ~/.keras/models/并在載入模型時(shí)自動(dòng)載入
# VGG16，VGG19，ResNet50 默認(rèn)輸入尺寸是224x224
# InceptionV3， InceptionResNetV2 模型的默認(rèn)輸入尺寸是299x299

# 使用內(nèi)置的預(yù)訓(xùn)練模型的步驟
# step1 導(dǎo)入需要的模型
# step2 將需要識(shí)別的圖像數(shù)據(jù)轉(zhuǎn)換為矩陣（矩陣的大小需要根據(jù)模型的不同而定）
# step3 將圖像矩陣丟到模型里面進(jìn)行預(yù)測
# -------------------------------------------------------
# step1 
import cv2
import numpy as np
from getConfig import getOption
from keras.applications import VGG16
from keras.applications import VGG19
from keras.applications import ResNet50
from keras.applications import InceptionV3
from keras.applications import InceptionResNetV2
from keras.applications import imagenet_utils
from keras.applications.imagenet_utils import decode_predictions
from keras.preprocessing.image import load_img
from keras.preprocessing.image import img_to_array
from keras.applications.inception_v3 import preprocess_input

class ImageTools(object):
  """
  使用keras預(yù)訓(xùn)練模型進(jìn)行圖像識(shí)別
  """
  def __init__(self, img, model, w):
    self.image = img
    self.model = model
    self.weight = w
    
  # step2
  def image2matrix(self, img):
    """
    將圖像轉(zhuǎn)為矩陣
    """
    image = cv2.imread(img)
    image = cv2.resize(image, self.dim)
    image = img_to_array(image)
    image = np.expand_dims(image, axis=0)
    return image

  @property
  def dim(self):
    """
    圖像矩陣的維度
    """
    if self.model in ["inceptionv3", "inceptionresnetv2"]:
      shape = (299, 299)
    else:
      shape = (224, 224)

    return shape

  @property
  def Model(self):
    """
    模型
    """
    models = {
      "vgg16": VGG16,
      "vgg19": VGG19,
      "resnet50": ResNet50,
      "inceptionv3": InceptionV3,
      "inceptionresnetv2": InceptionResNetV2
    }
    return models[self.model]

  # step3 
  def prediction(self):
    """
    預(yù)測
    """
    model = self.Model(weights=self.weight)
    if self.model in ["inceptionv3", "inceptionresnetv2"]:
      preprocess = preprocess_input(self.image2matrix(self.image))
    else:
      preprocess = imagenet_utils.preprocess_input(self.image2matrix(self.image))

    predict = model.predict(preprocess)

    decode_predict = decode_predictions(predict)

    for (item, (imgId, imgLabel, proba)) in enumerate(decode_predict[0]):
      print("{}, {}, {:.2f}%".format(item + 1, imgLabel, proba * 100))


if __name__ == "__main__":
  image = getOption("image", "image_path")
  model = getOption("model", "model")
  weight = getOption("weights", "weight")
  tools = ImageTools(image, model, weight)
  tools.prediction()