完整工程，deeplab v3+(tensorflow)代碼全理解及其運行過程，長期更新

前提：ubuntu+tensorflow-gpu+python3.6

各種環境提前配好

1.下載工程源碼

網址：https://github.com/tensorflow/models

下載時會遇到速度過慢或中間因為網絡錯誤停止，可以換移動網絡或者用迅雷下載。

2.測試環境

先添加slim路徑，每次打開terminal都要加載路徑

# From tensorflow/models/research/
export PYTHONPATH=$PYTHONPATH:`pwd`:`pwd`/slim

運行測試1

# From tensorflow/models/research/
python deeplab/model_test.py

測試2

# From tensorflow/models/research/deeplab
sh local_test.sh

3.處理數據標簽

處理標簽為單通道，運行下面代碼即可

import numpy as np
from PIL import Image
from keras.preprocessing.image import load_img, img_to_array
import os
 
classes = ['background', 'aeroplane', 'bicycle', 'bird', 'boat',
           'bottle', 'bus', 'car', 'cat', 'chair', 'cow', 'dining table',
           'dog', 'horse', 'motorbike', 'person', 'potted plant',
           'sheep', 'sofa', 'train', 'tv/monitor']
 
colormap = [[0, 0, 0], [128, 0, 0], [0, 128, 0], [128, 128, 0], [0, 0, 128],
            [128, 0, 128], [0, 128, 128], [128, 128, 128], [64, 0, 0], [192, 0, 0],
            [64, 128, 0], [192, 128, 0], [64, 0, 128], [192, 0, 128],
            [64, 128, 128], [192, 128, 128], [0, 64, 0], [128, 64, 0],
            [0, 192, 0], [128, 192, 0], [0, 64, 128]]
 
# 利用下面的代碼，將標注的圖片轉換為單通道的label圖像
cm2lbl = np.zeros(256**3)
for i, cm in enumerate(colormap):
    cm2lbl[(cm[0]*256+cm[1])*256+cm[2]] = i
 
 
def image2label(im):
    # 輸入為標記圖像的矩陣，輸出為單通道映射的label圖像
    data = im.astype('int32')
    idx = (data[:, :, 0]*256+data[:, :, 1])*256+data[:, :, 2]
    return np.array(cm2lbl[idx])
 
 
def change_label(label_url, label_name):
 
    label_img = load_img(label_url)
    label_img = img_to_array(label_img)
    label_img = image2label(label_img)  # 將圖片映射為單通道數據
    print(np.max(label_img))
 
    label_single = Image.fromarray(label_img)
    label_single = label_single.convert('L')
 
    save_path = './datasets/VOC2012/Label'
    save_path = os.path.join(save_path, label_name)  # 確定保存路徑及名稱
    label_single.save(save_path)
 
 
val_file_path = './datasets/VOC2012/ImageSets/trainval.txt'  # 文件名存放路徑
label_file_path = './datasets/VOC2012/SegmentationClass'  # 原label存放路徑
 
with open(val_file_path, 'r') as f:
    file_names = f.readlines()
    count = 0
    for name in file_names:
        count += 1
        name = name.strip('\n')  # 去掉換行符
        label_name = name + '.png'  # label文件名
        label_url = os.path.join(label_file_path, label_name)
        print('這是第 %s 張' % count)
        print(label_url)
        change_label(label_url, label_name)

4.運行build_voc2012_data.py 生成 .tfrecord數據

 

5. 運行train.py訓練模型