predict.py

from keras.models import load_model
from PIL import Image
import numpy as np
import os

labels = ['0','1','2','3','4','5','6','7','8','9','A','B','C','D','E','F','G','H','J','K','L','M','N','P','Q','R','S','T','U','V','W','X','Y','Z']  #標籤值
datan = 0  #車牌文字數,即檔案數
for fname in os.listdir('cropMono'):
    if os.path.isfile(os.path.join('cropMono', fname)):
        datan += 1
tem_data = []
for index in range(1, (datan+1)):  #讀取預測資料
    tem_data.append((np.array(Image.open("cropMono/" + str(index) +".jpg")))/255.0)
real_data = np.stack(tem_data)  #(6,38,18)
real_data1 = np.expand_dims(real_data, axis=3)  #轉換為(6,38,18,1)
model = load_model("carplate_model.hdf5")  #讀取模型
predictions = model.predict_classes(real_data1)  #預測資料
print('車牌號碼為：')
for i in range(len(predictions)):  #顯示結果
    print(labels[int(predictions[i])], end='')