【目标检测】keras-yolo3-tiny训练自己的数据集（目标：人&车辆）

浅浅的花香味﹌ 2022-12-06 04:03 240阅读 0赞

相关内容：[【目标检测】基于YOLOv3的海上船舶目标检测分类（Tensorflow/keras）][YOLOv3_Tensorflow_keras]

--------------------

### 目录 ###

*  一、环境配置
 *   *  1.1 配置GPU环境。
     *  1.2 虚拟环境与依赖
 *  二、数据集制作
 *   *  2.1 数据集制作
     *  2.2 生成ImageSets
 *  三、代码
 *   *  3.1 下载源码
     *  3.2 修改yolov3-tiny.cfg
     *  3.3 转换权重文件
     *  3.4 修改voc\_annotation
     *  3.5 修改yolo.py
     *  3.6 修改train.py
 *  四、测试
 *   *  4.1 测试集检测
     *  4.2 计算mAP

# 一、环境配置 #

## 1.1 配置GPU环境。 ##

使用GPU环境加快训练速度。

[【TensorFlow】Window10搭建GPU环境（CUDA、cuDNN）][TensorFlow_Window10_GPU_CUDA_cuDNN]

## 1.2 虚拟环境与依赖 ##

使用Anaconda创建虚拟环境，并安装相关依赖。

[Anaconda Prompt 常用命令][Anaconda Prompt]

**计算机环境**：Win10 + Python3.5 + cuda10.2

**主要依赖**：

tensorflow-gpu 1.10.0
    keras-gpu 2.2.2
    opencv
    pillow
    numpy
    matplotlib

# 二、数据集制作 #

## 2.1 数据集制作 ##

由于收集数据集并标准过于麻烦，所以采用现有数据集进行训练。

使用[Visual Object Classes Challenge 2012 (VOC2012)][Visual Object Classes Challenge 2012 _VOC2012]中的部分类别作为项目的数据集。

共包含20类目标，总计17125张图片。

该项目包含以下三个类别：`person`、`car`、`bus`。

关于数据集的制作、提取：[【python】数据集修改：移除和修改xml类别][python_xml]

## 2.2 生成ImageSets ##

制作好的数据集放在如下目录内：  
![在这里插入图片描述][20200912144729278.png_pic_left]

`jpg`文件在`JPEGImages`中；  
`xml`文件在`Annotations`中；  
`ImageSets`文件夹内新建`Main`备用。

使用代码`make_main_txt.py`在ImgeSets/Main目录下生成`train.txt`, `trainval.txt`, `val.txt`, `test.txt`。

该代码存放的目录如下图，`make_main_txt.py`与`ImageSets`同级。

# -*- coding:utf-8 -*-
    import os
    import random
    
    trainval_percent = 0.2    # test和val所占的比例，（1-trainval_percent）是训练集的比例
    train_percent = 0.8       # 其中在train时使用的test所占的比例
    
    xmlfilepath = 'Annotations'
    txtsavepath = 'ImageSets\Main'
    total_xml = os.listdir(xmlfilepath)
    
    num = len(total_xml)     # 图片数量
    list = range(num)
    tv = int(num * trainval_percent)  # trainval的数量
    tr = int(tv * train_percent)      # trainval中train的数量
    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()

# 三、代码 #

## 3.1 下载源码 ##

源码：[https://github.com/qqwweee/keras-yolo3][https_github.com_qqwweee_keras-yolo3]

## 3.2 修改yolov3-tiny.cfg ##

修改以下四部分：  
![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L1NlZV9TdGFy_size_16_color_FFFFFF_t_70_pic_left]  
![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L1NlZV9TdGFy_size_16_color_FFFFFF_t_70_pic_left 1]

filters = 3\*(类别数+5)  
classes = 类别数

当类别数量为2时，filters=21，classes=2

yolov3-tiny.cfg 注释参考：[https://blog.csdn.net/weixin\_44152895/article/details/106570976][https_blog.csdn.net_weixin_44152895_article_details_106570976]

## 3.3 转换权重文件 ##

在[YOLOv3][]下载[yolov3-tiny.weights][]。

使用命令`python convert.py yolov3-tiny.cfg yolov3-tiny.weights model_data/tiny_yolo_weights.h5`将`.weigth`文件转换为`.h5`文件。

## 3.4 修改voc\_annotation ##

# -*- coding:utf-8 -*-
    import xml.etree.ElementTree as ET
    from os import getcwd
    
    sets = ['train', 'val', 'test']
    classes = ["person", "vehicle"]
    
    
    def convert_annotation(image_id, list_file):
        print(image_id)
        in_file = open('dataset/Annotations/%s.xml' % image_id,encoding='utf-8')
        tree = ET.parse(in_file)
        root = tree.getroot()
    
        for obj in root.iter('object'):
            cls = obj.find('name').text
            if cls not in classes:
                continue
            cls_id = classes.index(cls)
            xmlbox = obj.find('bndbox')
            b = (int(xmlbox.find('xmin').text), int(xmlbox.find('ymin').text), int(xmlbox.find('xmax').text), int(xmlbox.find('ymax').text))
            list_file.write(" " + ",".join([str(a) for a in b]) + ',' + str(cls_id))
    
    wd = getcwd()
    
    for image_set in sets:
        image_ids = open('dataset/ImageSets/Main/%s.txt' % image_set).read().strip().split()
        print(image_ids)
        list_file = open('dataset/%s.txt' % image_set, 'w')
    
        for image_id in image_ids:
            list_file.write('dataset/JPEGImages/%s.jpg' % image_id)
            convert_annotation(image_id, list_file)
            list_file.write('\n')
        list_file.close()

运行代码将生产以下三个文件：  
![在这里插入图片描述][20200912145248381.png_pic_left]

## 3.5 修改yolo.py ##

修改yolo.py的默认项设置。

class YOLO(object):
        _defaults = { 
            "model_path": 'model_data/tiny_yolo_weights.h5',
            "anchors_path": 'model_data/tiny_yolo_anchors.txt',
            "classes_path": 'model_data/my_classes.txt',
            "score" : 0.3,
            "iou" : 0.45,
            "model_image_size" : (416, 416),
            "gpu_num" : 1,
        }

## 3.6 修改train.py ##

17-20行的路径设置：  
![在这里插入图片描述][20200912150212107.png_pic_left]  
52行的`if True`改成`if False`，因为这部分的训练效果不好，使用下一部分的训练。也就是从70行开始。

修改`batch_size`、`epochs`。

显存较小就讲`bathsize`设置的小一点。  
![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L1NlZV9TdGFy_size_16_color_FFFFFF_t_70_pic_left 2]  
最后，运行main.py。

# 四、测试 #

## 4.1 测试集检测 ##

创建main\_yolo.py，依照yolo.py进行修改。修改目录信息即可运行。可生成result文件夹，里面包含对测试集的检测结果。  
![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L1NlZV9TdGFy_size_16_color_FFFFFF_t_70_pic_left 3]

**main\_yolo.py**

# -*- coding: utf-8 -*-
    """ Class definition of YOLO_v3 style detection model on image and video """
    
    import colorsys
    import os
    import time
    from timeit import default_timer as timer
    
    import numpy as np
    from keras import backend as K
    from keras.models import load_model
    from keras.layers import Input
    from PIL import Image, ImageFont, ImageDraw
    
    from yolo3.model import yolo_eval, yolo_body, tiny_yolo_body
    from yolo3.utils import letterbox_image
    import os
    from keras.utils import multi_gpu_model
    
    
    dir_project = os.getcwd()  # 获取当前目录
    
    # 创建创建一个存储检测结果的dir
    result_path = './result'
    if not os.path.exists(result_path):
        os.makedirs(result_path)
    
    # result如果之前存放的有文件，全部清除
    for i in os.listdir(result_path):
        path_file = os.path.join(result_path, i)
        if os.path.isfile(path_file):
            os.remove(path_file)
    
    # 创建一个记录检测结果的文件
    txt_path = result_path + '/result.txt'
    file = open(txt_path, 'w')
    
    
    class YOLO(object):
        _defaults = { 
            "model_path": 'logs/000/trained_weights_final.h5',
            "anchors_path": 'model_data/tiny_yolo_anchors.txt',
            "classes_path": 'model_data/my_classes.txt',
            "score" : 0.3,
            "iou" : 0.45,
            "model_image_size" : (416, 416),
            "gpu_num" : 1,
        }
    
        @classmethod
        def get_defaults(cls, n):
            if n in cls._defaults:
                return cls._defaults[n]
            else:
                return "Unrecognized attribute name '" + n + "'"
    
        def __init__(self, **kwargs):
            self.__dict__.update(self._defaults) # set up default values
            self.__dict__.update(kwargs) # and update with user overrides
            self.class_names = self._get_class()
            self.anchors = self._get_anchors()
            self.sess = K.get_session()
            self.boxes, self.scores, self.classes = self.generate()
    
        def _get_class(self):
            classes_path = os.path.expanduser(self.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(self):
            anchors_path = os.path.expanduser(self.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 generate(self):
            model_path = os.path.expanduser(self.model_path)
            assert model_path.endswith('.h5'), 'Keras model or weights must be a .h5 file.'
    
            # Load model, or construct model and load weights.
            num_anchors = len(self.anchors)
            num_classes = len(self.class_names)
            is_tiny_version = num_anchors==6 # default setting
            try:
                self.yolo_model = load_model(model_path, compile=False)
            except:
                self.yolo_model = tiny_yolo_body(Input(shape=(None,None,3)), num_anchors//2, num_classes) \
                    if is_tiny_version else yolo_body(Input(shape=(None,None,3)), num_anchors//3, num_classes)
                self.yolo_model.load_weights(self.model_path) # make sure model, anchors and classes match
            else:
                assert self.yolo_model.layers[-1].output_shape[-1] == \
                    num_anchors/len(self.yolo_model.output) * (num_classes + 5), \
                    'Mismatch between model and given anchor and class sizes'
    
            print('{} model, anchors, and classes loaded.'.format(model_path))
    
            # Generate colors for drawing bounding boxes.
            hsv_tuples = [(x / len(self.class_names), 1., 1.)
                          for x in range(len(self.class_names))]
            self.colors = list(map(lambda x: colorsys.hsv_to_rgb(*x), hsv_tuples))
            self.colors = list(
                map(lambda x: (int(x[0] * 255), int(x[1] * 255), int(x[2] * 255)),
                    self.colors))
            np.random.seed(10101)  # Fixed seed for consistent colors across runs.
            np.random.shuffle(self.colors)  # Shuffle colors to decorrelate adjacent classes.
            np.random.seed(None)  # Reset seed to default.
    
            # Generate output tensor targets for filtered bounding boxes.
            self.input_image_shape = K.placeholder(shape=(2, ))
            if self.gpu_num>=2:
                self.yolo_model = multi_gpu_model(self.yolo_model, gpus=self.gpu_num)
            boxes, scores, classes = yolo_eval(self.yolo_model.output, self.anchors,
                    len(self.class_names), self.input_image_shape,
                    score_threshold=self.score, iou_threshold=self.iou)
            return boxes, scores, classes
    
        def detect_image(self, image):
            start = timer()
    
            if self.model_image_size != (None, None):
                assert self.model_image_size[0]%32 == 0, 'Multiples of 32 required'
                assert self.model_image_size[1]%32 == 0, 'Multiples of 32 required'
                boxed_image = letterbox_image(image, tuple(reversed(self.model_image_size)))
            else:
                new_image_size = (image.width - (image.width % 32),
                                  image.height - (image.height % 32))
                boxed_image = letterbox_image(image, new_image_size)
            image_data = np.array(boxed_image, dtype='float32')
    
            print(image_data.shape)
            image_data /= 255.
            image_data = np.expand_dims(image_data, 0)  # Add batch dimension.
    
            out_boxes, out_scores, out_classes = self.sess.run(
                [self.boxes, self.scores, self.classes],
                feed_dict={ 
                    self.yolo_model.input: image_data,
                    self.input_image_shape: [image.size[1], image.size[0]],
                    K.learning_phase(): 0
                })
    
            print('Found {} boxes for {}'.format(len(out_boxes), 'img'))
    
            font = ImageFont.truetype(font='font/FiraMono-Medium.otf',
                        size=np.floor(3e-2 * image.size[1] + 0.5).astype('int32'))
            thickness = (image.size[0] + image.size[1]) // 300
    
            # # 保存框检测出的框的个数 (添加)
            # file.write('find ' + str(len(out_boxes)) + ' target(s) \n')
    
            for i, c in reversed(list(enumerate(out_classes))):
                predicted_class = self.class_names[c]
                box = out_boxes[i]
                score = out_scores[i]
    
                label = '{} {:.2f}'.format(predicted_class, score)
                draw = ImageDraw.Draw(image)
                label_size = draw.textsize(label, font)
    
                top, left, bottom, right = box
                top = max(0, np.floor(top + 0.5).astype('int32'))
                left = max(0, np.floor(left + 0.5).astype('int32'))
                bottom = min(image.size[1], np.floor(bottom + 0.5).astype('int32'))
                right = min(image.size[0], np.floor(right + 0.5).astype('int32'))
    
                # # 写入检测位置（添加）
                # file.write(
                # predicted_class + ' score: ' + str(score) + ' \nlocation: top: ' + str(top) + '、 bottom: ' + str(
                # bottom) + '、 left: ' + str(left) + '、 right: ' + str(right) + '\n')
    
                file.write(predicted_class + ' ' + str(score) + ' ' + str(left) + ' ' + str(top) + ' ' + str(right) + ' ' + str(bottom) + ';')
    
                print(label, (left, top), (right, bottom))
    
                if top - label_size[1] >= 0:
                    text_origin = np.array([left, top - label_size[1]])
                else:
                    text_origin = np.array([left, top + 1])
    
                # My kingdom for a good redistributable image drawing library.
                for i in range(thickness):
                    draw.rectangle(
                        [left + i, top + i, right - i, bottom - i],
                        outline=self.colors[c])
                draw.rectangle(
                    [tuple(text_origin), tuple(text_origin + label_size)],
                    fill=self.colors[c])
                draw.text(text_origin, label, fill=(0, 0, 0), font=font)
                del draw
            end = timer()
            print(end - start)
            return image
    
        def close_session(self):
            self.sess.close()
    
    
    
    def detect_video(yolo, video_path, output_path=""):
        import cv2
        vid = cv2.VideoCapture(video_path)
        if not vid.isOpened():
            raise IOError("Couldn't open webcam or video")
        video_FourCC    = int(vid.get(cv2.CAP_PROP_FOURCC))        # 获得视频编码MPEG4/H264
        video_fps       = vid.get(cv2.CAP_PROP_FPS)
        video_size      = (int(vid.get(cv2.CAP_PROP_FRAME_WIDTH)),
                            int(vid.get(cv2.CAP_PROP_FRAME_HEIGHT)))
        isOutput = True if output_path != "" else False
        if isOutput:
            print("!!! TYPE:", type(output_path), type(video_FourCC), type(video_fps), type(video_size))
            out = cv2.VideoWriter(output_path, video_FourCC, video_fps, video_size)
        accum_time = 0
        curr_fps = 0
        fps = "FPS: ??"
        prev_time = timer()
        while True:
            return_value, frame = vid.read()
            image = Image.fromarray(frame)           # 从array转换成image
            image = yolo.detect_image(image)
            result = np.asarray(image)
            curr_time = timer()
            exec_time = curr_time - prev_time
            prev_time = curr_time
            accum_time = accum_time + exec_time
            curr_fps = curr_fps + 1
            if accum_time > 1:
                accum_time = accum_time - 1
                fps = "FPS: " + str(curr_fps)
                curr_fps = 0
            cv2.putText(result, text=fps, org=(3, 15), fontFace=cv2.FONT_HERSHEY_SIMPLEX,
                        fontScale=0.50, color=(255, 0, 0), thickness=2)
            cv2.namedWindow("result", cv2.WINDOW_NORMAL)
            cv2.imshow("result", result)
            if isOutput:
                out.write(result)
            if cv2.waitKey(1) & 0xFF == ord('q'):
                break
        yolo.close_session()
    
    
    # 批量处理文件
    if __name__ == '__main__':
        if True:
            # 读取test文件
            with open("dataset/ImageSets/Main/test.txt", 'r') as f:  # 打开文件
                test_list = f.readlines()  # 读取文件
                test_list = [x.strip() for x in test_list if x.strip() != '']  # 去除/n
                # print(test_list)
    
            t1 = time.time()
            yolo = YOLO()
    
            for filename in test_list:
                image_path = 'dataset/JPEGImages/'+filename+'.jpg'
                portion = os.path.split(image_path)
                # file.write(portion[1]+' detect_result：\n')
                file.write(image_path + ' ')
                image = Image.open(image_path)
                image_mAP_save_path = dir_project + '/mAP/input/images-optional/'
                image.save(image_mAP_save_path + filename + '.jpg')
                r_image = yolo.detect_image(image)
                file.write('\n')
                #r_image.show() 显示检测结果
                image_save_path = './result/result_'+portion[1]
                print('detect result save to....:'+image_save_path)
                r_image.save(image_save_path)
    
            time_sum = time.time() - t1
            # file.write('time sum: '+str(time_sum)+'s')
            print('time sum:',time_sum)
            file.close()
            yolo.close_session()
    
        if False:
            img_path = 'dataset/JPEGImages_resize'
            result_path = 'dataset/image_result'
            for file in os.listdir(img_path):
                file_path = os.path.join(img_path, file)
                image = Image.open(file_path)
                r_image = yolo.detect_image(image)
    
                image_save_path = os.path.join(result_path, file)
                print('detect result save to: ' + image_save_path)
                r_image.save(image_save_path)

## 4.2 计算mAP ##

mAP计算：[【目标检测】kera-yolo3模型计算mAP][kera-yolo3_mAP]  
mAP源码：[https://github.com/Cartucho/mAP][https_github.com_Cartucho_mAP]

修改mAP源码，针对keras-yolov3的检测结果计算mAP。

[YOLOv3_Tensorflow_keras]: https://blog.csdn.net/See_Star/article/details/105820181
[TensorFlow_Window10_GPU_CUDA_cuDNN]: https://blog.csdn.net/See_Star/article/details/103141180
[Anaconda Prompt]: https://blog.csdn.net/See_Star/article/details/105028411
[Visual Object Classes Challenge 2012 _VOC2012]: http://host.robots.ox.ac.uk/pascal/VOC/voc2012/index.html
[python_xml]: https://blog.csdn.net/See_Star/article/details/108549326
[20200912144729278.png_pic_left]: /images/20221123/0860078836bb4d6f8e592ffc76e4e4ca.png
[https_github.com_qqwweee_keras-yolo3]: https://github.com/qqwweee/keras-yolo3
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L1NlZV9TdGFy_size_16_color_FFFFFF_t_70_pic_left]: /images/20221123/4800147b9d6548b79214690b9947e35b.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L1NlZV9TdGFy_size_16_color_FFFFFF_t_70_pic_left 1]: /images/20221123/0b0fe31d76b24c178b783cfcd5237a8c.png
[https_blog.csdn.net_weixin_44152895_article_details_106570976]: https://blog.csdn.net/weixin_44152895/article/details/106570976
[YOLOv3]: https://pjreddie.com/darknet/yolo/
[yolov3-tiny.weights]: https://pjreddie.com/media/files/yolov3-tiny.weights
[20200912145248381.png_pic_left]: /images/20221123/3497e3ea80cb4d919a20952f7336df4e.png
[20200912150212107.png_pic_left]: /images/20221123/888ddfecd0a44587893c52775ea60953.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L1NlZV9TdGFy_size_16_color_FFFFFF_t_70_pic_left 2]: /images/20221123/898f08d7daab483495b6805fd8b61e29.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L1NlZV9TdGFy_size_16_color_FFFFFF_t_70_pic_left 3]: /images/20221123/312eaac7122c4845ab6cafac454339ad.png
[kera-yolo3_mAP]: https://blog.csdn.net/See_Star/article/details/104430955
[https_github.com_Cartucho_mAP]: https://github.com/Cartucho/mAP