tensorflow环境下训练keras-yolov3制作自己的数据集全过程

keras-yolov3资源包下载：

地址：https://pan.baidu.com/s/1MT-UqJM669kqdRIrEkKQ8g

提取码：gbjk

环境：

i7-9750H；

无N卡；

tensorflow-cpu；

tensorflow第三方库：keras、numpy、pillow等

步骤：

1、最终的keras-yolov3new文件结构如下：

2、建立VOC数据集合。建立VOCdevkit文件，结构如下：

将自己搜集到的图片数据存储在JPEGImages文件下；

利用LabelImg工具将图片标注成xml格式存储在Annotations文件下；

新建test.py,代码如下：

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()

划分训练集和测试集，存储在ImageSets文件下：

3、回到目录：D:\python\keras-yolo3new，修改voc_annotation.py的程序，将class修改成自己的类别：

修改完成后，在D:\python\keras-yolo3new目录下生成3个2007开头的txt文件：007_train.txt，2007_test.txt，2007_val.txt。删除这3个txt文件文件名中的“2007_”这部分，就变成了：train.txt，test.txt，val.txt

 

4、修改参数文件yolo3.cfg
3处yolo，注意，filters在yolo的上边，其数值为3*(5+类数)，classes和random在yolo下边。
在keras-yolo3-master目录下打开yolo3.cfg，搜索yolo，会发现有3处包含yolo，
每个地方都要改3处，

filters：3*（5+len（classes））；       # 这个在yolo    上方的代码块里
classes: len(classes) = 1， #注意你有几类   这个在yolo的代码块里
random：原来是1，显存小改为0  #在yolo里

 

5、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:

6、修改train.py

"""
Retrain the YOLO model for your own dataset.
"""
import os
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

from yolo3.model import preprocess_true_boxes, yolo_body, tiny_yolo_body, yolo_loss
from yolo3.utils import get_random_data

os.environ["CUDA_VISIBLE_DEVICES"] = "-1" 
def _main():
    annotation_path = 'D:/python/keras-yolo3new/train.txt'
    log_dir = 'D:/python/keras-yolo3new/logs/000/'
    classes_path = 'D:/python/keras-yolo3new/model_data/voc_classes.txt'
    anchors_path = 'D:/python/keras-yolo3new/model_data/yolo_anchors.txt'
    class_names = get_classes(classes_path)
    anchors = get_anchors(anchors_path)
    input_shape = (416,416) # multiple of 32, hw
    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/'):
    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, save_best_only=True, period=1)
    batch_size = 8#按照自己的显存修改大小
    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))

    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=300,#修改迭代次数
            initial_epoch=0)
    model.save_weights(log_dir + 'trained_weights.h5')#若后续要转换成pb文件，改成save()

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()    

7、运行yolo.py

修改：

class YOLO(object):
    _defaults = {
        "model_path": 'logs/000/trained_weights_final.h5',
        "anchors_path": 'model_data/yolo_anchors.txt',
        "classes_path": 'model_data/coco_classes.txt',
        "score" : 0.35,
        "iou" : 0.8,
        "model_image_size" : (416, 416),
        "gpu_num" : 0,
    }

若是没有出现框，可以修改score和iou的数值

最后加上：

if __name__ == '__main__':    
    yolo=YOLO()    
    path = 'D:/python/keras-yolo3new/VOCdevkit/VOC2007/JPEGImages/000001.jpg'    
    try:        
        image = Image.open(path)    
    except:        
        print('Open Error! Try again!')    
    else:        
        r_image= yolo.detect_image(image)        
        r_image.show()     
    yolo.close_session()

小编最后测试的效果不太好，估计是某些参数没修改好，loss很大，估计是训练次数太少或者是图片数量太少，希望大神指教