【目标检测】Keras/Tensorflow+python+yolo3,YOLO_tensorflow-master

一、YOLO_tensorflow-master直接测试[转]

参考博文：https://blog.csdn.net/c20081052/article/details/80260726

0.运行环境：

Win7 +  python3.5+  PyCharm 2018 x64+TensorFlow1.6.0+12G内存

1. 代码下载

链接是：https://github.com/hizhangp/yolo_tensorflow

下载代码YOLO_tensorflow-master.zip解压后，新建两级文件夹data/weight,用于存放YOLO模型。

2.YOLO权重文件下载，

链接: https://pan.baidu.com/s/1htt9YBE 密码: ehw2
下载后的weights文件YOLO_small.ckpt放在data/weight文件夹下

3.直接运行test.py

只用修改测试图片的名字

---

二、Keras/Tensorflow+python+yolo3训练自己的数据集

参考博文：https://blog.csdn.net/Patrick_Lxc/article/details/80615433

该程序环境搭建：https://blog.csdn.net/imwaters/article/details/84938839

 代码下载：https://github.com/qqwweee/keras-yolo3

 

2.制作自己的VOC数据集

2.1首先新建一个名字为VOC2007的文件夹，在该文件夹下再分别创建Annotations、ImageSets、JPEGImages三个子文件夹

在ImageSets文件夹下创建名为Main的文件夹。【参考链接】

2.2 将训练用的图片重命名为VOC形式0000X-00XXX后保存到JPEGImage文件夹中。【使用代码重命名图片】

2.3 安装像标注工具LabelImage。

网盘下载：https://pan.baidu.com/s/1NEAyAZxP7mfEhDG5FXXuGQ

下载图像标注工具LabelImage【官方下载】,标记生成xml文件。

• 在“predefined_classes.txt”修改默认标签名称，重启LabelImage
• 点击Open Dir 选择 JPEGImage 文件夹
• 点击Change Save Dir  选择 保存在Annotations 文件夹
• 点击Create RectBox 进行标签工作。快捷键 “w” ，标注完点“Save”
• 点击Next Image 进行下一张标签，快捷键 “d”，上一张图 ，快捷键 “a”

 

3.生成训练数据集及标签

3.1 代码生成ImageSets/Main/4个文件，在VOC2007文件夹下，新建并运行下列代码

4个txt文件，其中test.txt是测试集，train.txt是训练集，val.txt是验证集，trainval.txt是训练和验证集。【原文链接】

import os
import random

trainval_percent = 0.1
train_percent = 0.9
xmlfilepath = 'Annotations'
txtsavepath = 'ImageSets\Main'
total_xml = os.listdir(xmlfilepath)

num = len(total_xml)
list = range(num)
tv = int(num * trainval_percent)
tr = int(tv * train_percent)
trainval = random.sample(list, tv)
train = random.sample(trainval, tr)

ftrainval = open('ImageSets/Main/trainval.txt', 'w')
ftest = open('ImageSets/Main/test.txt', 'w')
ftrain = open('ImageSets/Main/train.txt', 'w')
fval = open('ImageSets/Main/val.txt', 'w')

for i in list:
    name = total_xml[i][:-4] + '\n'
    if i in trainval:
        ftrainval.write(name)
        if i in train:
            ftest.write(name)
        else:
            fval.write(name)
    else:
        ftrain.write(name)

ftrainval.close()
ftrain.close()
fval.close()
ftest.close()

3.2 生成yolo3所需的train.txt,val.txt,test.txt
  VOC2007数据格式数据集制作完成，需要转换成keras-tensorflow使用的格式。

  修改并运行voc_annotation.py 

    运行之后，会在主目录下多生成三个txt文件，

重命名，手动删除2007_,

3.3 修改model_data下的文件，放入你的类别，coco_classes,voc_classes这两个文件都需要修改。

   

 

4.修改yolo3.cfg配置

作者：Patrick_Lxc 
原文：https://blog.csdn.net/Patrick_Lxc/article/details/80615433

IDE里直接打开cfg文件，ctrl+f搜 yolo, 总共会搜出3个含有yolo的地方！！

    每个地方都要改3处，filters：3*（5+len（classes））；

                                    classes: len(classes) = 5，这里本人有三个

                                    random：原来是1，显存小改为0    

5.修改train.py(可直接复制下列代码) ，训练自己的数据集

其中，文件目录，批次大小，train与val级大小，迭代次数有很大区别（epochs）

 

import numpy as np
import keras.backend as K
from keras.layers import Input, Lambda
from keras.models import Model
from keras.callbacks import TensorBoard, ModelCheckpoint, EarlyStopping
import os
from yolo3.model import preprocess_true_boxes, yolo_body, tiny_yolo_body, yolo_loss
from yolo3.utils import get_random_data
import os
os.environ['TF_MIN_GPU_MULTIPROCESSOR_COUNT'] = '1'

def _main():
    annotation_path = 'train.txt'
    log_dir = './logs/'
    classes_path = 'model_data/voc_classes_181220.txt'
    anchors_path = 'model_data/yolo_anchors.txt'
    class_names = get_classes(classes_path)
    anchors = get_anchors(anchors_path)
    input_shape = (416,416) # multiple of 32,
    model = create_model(input_shape, anchors, len(class_names) )
    train(model, annotation_path, input_shape, anchors, len(class_names), log_dir=log_dir)

def train(model, annotation_path, input_shape, anchors, num_classes, log_dir='logs/',exist_h5='logs/guilin_181220.h5'):
    model.compile(optimizer='adam', loss={
        'yolo_loss': lambda y_true, y_pred: y_pred})
    logging = TensorBoard(log_dir=log_dir)
    checkpoint = ModelCheckpoint(log_dir + "ep{epoch:03d}-loss{loss:.3f}-val_loss{val_loss:.3f}.h5",
        # monitor='val_loss', save_weights_only=True, verbose = 1, save_best_only=True, period=1)
        monitor='val_loss', save_weights_only=True, save_best_only=True, period=1)

    #批次大小是重要数据
    batch_size =5
    val_split = 0.1
    with open(annotation_path) as f:
        lines = f.readlines()
    np.random.shuffle(lines)
    num_val = int(len(lines)*val_split)
    num_train = len(lines) - num_val
    print('Train on {} samples, val on {} samples, with batch size {}.'.format(num_train, num_val, batch_size))

    if os.path.exists(exist_h5):
        model.load_weights(exist_h5)
        # 若成功加载前面保存的参数，输出下列信息
        print("------------checkpoint_loaded--------")

    model.fit_generator(data_generator_wrap(lines[:num_train], batch_size, input_shape, anchors, num_classes),
            steps_per_epoch=max(1, num_train//batch_size),
            validation_data=data_generator_wrap(lines[num_train:], batch_size, input_shape, anchors, num_classes),
            validation_steps=max(1, num_val//batch_size),
            epochs=500,
            #epochs=500,
            initial_epoch=0,
            #保存训练过程
            callbacks=[checkpoint, TensorBoard(log_dir=log_dir)])
    #训练好后的名字
    model.save_weights(log_dir + 'guilin_181220.h5')

def get_classes(classes_path):
    with open(classes_path) as f:
        class_names = f.readlines()
    class_names = [c.strip() for c in class_names]
    return class_names

def get_anchors(anchors_path):
    with open(anchors_path) as f:
        anchors = f.readline()
    anchors = [float(x) for x in anchors.split(',')]
    return np.array(anchors).reshape(-1, 2)

def create_model(input_shape, anchors, num_classes, load_pretrained=False, freeze_body=False,
            weights_path='model_data/yolo_weights.h5'):
    K.clear_session() # get a new session
    image_input = Input(shape=(None, None, 3))
    h, w = input_shape
    num_anchors = len(anchors)
    y_true = [Input(shape=(h//{0:32, 1:16, 2:8}[l], w//{0:32, 1:16, 2:8}[l], \
        num_anchors//3, num_classes+5)) for l in range(3)]

    model_body = yolo_body(image_input, num_anchors//3, num_classes)
    print('Create YOLOv3 model with {} anchors and {} classes.'.format(num_anchors, num_classes))

    if load_pretrained:
        model_body.load_weights(weights_path, by_name=True, skip_mismatch=True)
        print('Load weights {}.'.format(weights_path))
        if freeze_body:
            # Do not freeze 3 output layers.
            num = len(model_body.layers)-7
            for i in range(num): model_body.layers[i].trainable = False
            print('Freeze the first {} layers of total {} layers.'.format(num, len(model_body.layers)))

    model_loss = Lambda(yolo_loss, output_shape=(1,), name='yolo_loss',
        arguments={'anchors': anchors, 'num_classes': num_classes, 'ignore_thresh': 0.5})(
        [*model_body.output, *y_true])
    model = Model([model_body.input, *y_true], model_loss)
    return model
def data_generator(annotation_lines, batch_size, input_shape, anchors, num_classes):
    n = len(annotation_lines)
    np.random.shuffle(annotation_lines)
    i = 0
    while True:
        image_data = []
        box_data = []
        for b in range(batch_size):
            i %= n
            image, box = get_random_data(annotation_lines[i], input_shape, random=True)
            image_data.append(image)
            box_data.append(box)
            i += 1
        image_data = np.array(image_data)
        box_data = np.array(box_data)
        y_true = preprocess_true_boxes(box_data, input_shape, anchors, num_classes)
        yield [image_data, *y_true], np.zeros(batch_size)

def data_generator_wrap(annotation_lines, batch_size, input_shape, anchors, num_classes):
    n = len(annotation_lines)
    if n==0 or batch_size<=0: return None
    return data_generator(annotation_lines, batch_size, input_shape, anchors, num_classes)

if __name__ == '__main__':
    _main()

 

 

图像标注工具labelImg安装方法

1.参考百度链接。【失败】

2.官方github安装 https://github.com/tzutalin/labelImg     [成功]

直接下载windows安装包，

 windows下 labelImgexe闪退解决方案【转】

 https://blog.csdn.net/csdn_zhishui/article/details/84583888

删除 C:\Users\YourAccount\..labelImgSettings.pkl，重新打开labelImg，问题解决。