全球人工智能技术创新大赛【热身赛一】布匹疵点智能识别Baseline

比赛地址：https://tianchi.aliyun.com/competition/entrance/531864/introduction?spm=5176.12281949.1003.16.493e8f15PPTpkV
baseline地址：https://github.com/datawhalechina/team-learning-cv/tree/master/DefectDetection (我自己没有在baseline上跑通，后来重新clone yolov5官方镜像，在文中将会提到)
yolov5官方地址：https://github.com/ultralytics/yolov5
yolo5中文汉化地址：https://github.com/wudashuo/yolov5

赛题背景

在布匹的实际生产过程中，由于各方面因素的影响，会产生污渍、破洞、毛粒等瑕疵，为保证产品质量，需要对布匹进行瑕疵检测。布匹疵点检验是纺织行业生产和质量管理的重要环节，目前人工检测易受主观因素影响，缺乏一致性；并且检测人员在强光下长时间工作对视力影响极大。由于布匹疵点种类繁多、形态变化多样、观察识别难道大，导致布匹疵点智能检测是困扰行业多年的技术瓶颈。

近年来，人工智能和计算机视觉等技术突飞猛进，在工业质检场景中也取得了不错的成果。纺织行业迫切希望借助最先进的技术，实现布匹疵点智能检测。革新质检流程，自动完成质检任务，降低对大量人工的依赖，减少漏检发生率，提高产品的质量。

本赛场聚焦布匹疵点智能检测，要求选手研究开发高效可靠的计算机视觉算法，提升布匹疵点检验的准确度，降低对大量人工的依赖，提升布样疵点质检的效果和效率。要求算法既要检测布匹是否包含疵点，又要给出疵点具体的位置和类别，既考察疵点检出能力、也考察疵点定位和分类能力。

赛题数据

赛题组深入佛山南海纺织车间现场采集布匹图像，制作并发布大规模的高质量布匹疵点数据集，同时提供精细的标注来满足算法要求。大赛数据涵盖了纺织业中布匹的各类重要瑕疵，每张图片含一个或多种瑕疵。本次比赛主要使用花色布数据，约12000张。

数据示例
花色布数据包含原始图片、模板图片和瑕疵的标注数据。标注数据详细标注出疵点所在的具体位置和疵点类别，，数据示例如下。

全球人工智能技术创新大赛【热身赛一】布匹疵点智能识别Baseline_第1张图片

enter image description here

训练数据文件结构

我们将提供用于训练的图像数据和识别标签，文件夹结构：

|-- defect Images #存放有瑕疵的图像数据
|-- normal Images #存放无疵点的图像数据，jpeg编码图像文件
|-- Annotations #存放属性标签标注数据
|-- README.md #对数据的详细介绍

数据下载地址：
guangdong1_round2_train2_20191004_images.zip
guangdong1_round2_train2_20191004_Annotations.zip

代码运行

前提

将数据集下载后放到代码根目录下的train_data文件夹中，并解压
在yolov5 release下下载预训练好的权重

运行过程

python convertTrainLabel.py
python process_data_yolo.py
修改process_data_yolo.py
rain.sh文件中，第二步使用了process_data_yolo.py，源码中关于数据集存放位置存在问题，只写了val的处理，没写train的处理，所以生成的process_data文件夹中，只有val而没有train，训练时会报错。

所以不能直接用train.sh脚本，要顺序运行里面的命令，到第二步的时候，先执行一遍，如下图做修改后再执行一遍，从而把训练集和验证集都准备好。

全球人工智能技术创新大赛【热身赛一】布匹疵点智能识别Baseline_第2张图片

再次运行python process_data_yolo.py
rm -rf ./convertor
python train.py，在运行train.py的时候遇到了问题，我首先从官方仓库中下载了yolov5x.pt权重，但是在运行过程中报错AttributeError:Can't get attribute 'C3' on ,也就说说没有C3这个属性，官方Issue,两个建议一是安装完整的requirements.txt,尝试无果；二是说代码需要更新，我就重新clone了官方代码，运行OK，没有问题了，运行5个epoch，几分钟时间，速度就是很快，开心。
注意
1. convertTrainLabel和process_data_yolo代码其实这个时候用不到了，可以不拷贝过来，但是需要把最终生成的process_data目录放到yolov5代码根目录下
2. 修改data/coco128.yaml，作用为指定训练和验证集目录，指定类别数量和种类
```
# train and val datasets (image directory or *.txt file with image paths)
train: ./process_data/images/train/
val: ./process_data/images/val/
# number of classes
nc: 15
# class names
names: ['1', '2', '3', '4', '5', '6', '7', '8', '9', '10', '11', '12','13', '14', '15']
```

提交Docker

运行完以后就需要打包docker镜像了

将yolov5代码重新放到一个文件中中，如yolo，直接使用clone的也行，但是记得删除训练的图片，要不打包镜像很大的
修改dockerfile，内容如下，我为了打包的镜像小一点直接用的基础python3的镜像，这个是没有GPU的，不过只是测试了，有没有都没那么重要了

# Start FROM Nvidia PyTorch image https://ngc.nvidia.com/catalog/containers/nvidia:pytorch
FROM registry.cn-shanghai.aliyuncs.com/tcc-public/python:3

# Install linux packages
RUN apt update && apt install -y screen libgl1-mesa-glx



## 把当前文件夹里的文件构建到镜像的根目录下
ADD . /

## 指定默认工作目录为根目录（需要把run.sh和生成的结果文件都放在该文件夹下，提交后才能运行）
WORKDIR /

# Install python dependencies
RUN python -m pip install --upgrade pip
RUN pip install -i https://pypi.tuna.tsinghua.edu.cn/simple -r requirements.txt gsutil

## 镜像启动后统一执行 sh run.sh
CMD ["sh", "run.sh"]

修改requirements.txt

# pip install -r requirements.txt

# base ----------------------------------------
Cython
matplotlib>=3.2.2
numpy>=1.18.5
opencv-python>=4.1.2
Pillow
PyYAML>=5.3.1
scipy>=1.4.1
tensorboard>=2.2
torch>=1.7.0
torchvision>=0.8.1
tqdm>=4.41.0

# logging -------------------------------------
# wandb

# plotting ------------------------------------
seaborn>=0.11.0
pandas

# export --------------------------------------
# coremltools>=4.1
# onnx>=1.8.1
# scikit-learn==0.19.2  # for coreml quantization

# extras --------------------------------------
thop  # FLOPS computation
pycocotools>=2.0  # COCO mAP

创建run.sh

python getImage.py
python detect.py

创建getImage.py，因为天池的测试数据集都在tcdata下，而且一张图片一个文件夹，为了方便就将待检测都放到defect文件夹下

import os
import shutil as sh
defect_imgs = '/defect'
os.makedirs(defect_imgs)
path = '/tcdata/guangdong1_round2_testB_20191024'
folders = os.listdir(path)
for folder in folders:
    locations = os.path.join(path,folder,folder+'.jpg')
    if os.path.exists(locations):
        sh.copy(locations,defect_imgs)

修改detect.py，参照baseline的代码修改yolov5的detect代码，主要是为了保存需要提交的result.json,该文件需要注意的就是一个地方：参数weights需要将你生成的权重地址修改好

import argparse
import time
from pathlib import Path

import cv2
import torch
import torch.backends.cudnn as cudnn
from numpy import random
import os
import json
from models.experimental import attempt_load
from utils.datasets import LoadStreams, LoadImages
from utils.general import check_img_size, check_requirements, check_imshow, non_max_suppression, apply_classifier, \
    scale_coords, xyxy2xywh, strip_optimizer, set_logging, increment_path
from utils.plots import plot_one_box
from utils.torch_utils import select_device, load_classifier, time_synchronized


def detect(save_img=False):
    source, weights, view_img, save_txt, imgsz = opt.source, opt.weights, opt.view_img, opt.save_txt, opt.img_size
    webcam = source.isnumeric() or source.endswith('.txt') or source.lower().startswith(
        ('rtsp://', 'rtmp://', 'http://'))

    # Directories
    save_dir = Path(increment_path(Path(opt.project) / opt.name, exist_ok=opt.exist_ok))  # increment run
    (save_dir / 'labels' if save_txt else save_dir).mkdir(parents=True, exist_ok=True)  # make dir

    # Initialize
    set_logging()
    device = select_device(opt.device)
    half = device.type != 'cpu'  # half precision only supported on CUDA

    # Load model
    model = attempt_load(weights, map_location=device)  # load FP32 model
    stride = int(model.stride.max())  # model stride
    imgsz = check_img_size(imgsz, s=stride)  # check img_size
    if half:
        model.half()  # to FP16

    # Second-stage classifier
    classify = False
    if classify:
        modelc = load_classifier(name='resnet101', n=2)  # initialize
        modelc.load_state_dict(torch.load('weights/resnet101.pt', map_location=device)['model']).to(device).eval()

    # Set Dataloader
    vid_path, vid_writer = None, None
    if webcam:
        view_img = check_imshow()
        cudnn.benchmark = True  # set True to speed up constant image size inference
        dataset = LoadStreams(source, img_size=imgsz, stride=stride)
    else:
        save_img = True
        dataset = LoadImages(source, img_size=imgsz, stride=stride)

    # Get names and colors
    names = model.module.names if hasattr(model, 'module') else model.names
    colors = [[random.randint(0, 255) for _ in range(3)] for _ in names]

    # Run inference
    if device.type != 'cpu':
        model(torch.zeros(1, 3, imgsz, imgsz).to(device).type_as(next(model.parameters())))  # run once
    t0 = time.time()
    
    save_json = True
    result = []
    
    for path, img, im0s, vid_cap in dataset:
        img = torch.from_numpy(img).to(device)
        img = img.half() if half else img.float()  # uint8 to fp16/32
        img /= 255.0  # 0 - 255 to 0.0 - 1.0
        if img.ndimension() == 3:
            img = img.unsqueeze(0)

        # Inference
        t1 = time_synchronized()
        pred = model(img, augment=opt.augment)[0]

        # Apply NMS
        pred = non_max_suppression(pred, opt.conf_thres, opt.iou_thres, classes=opt.classes, agnostic=opt.agnostic_nms)
        t2 = time_synchronized()

        # Apply Classifier
        if classify:
            pred = apply_classifier(pred, modelc, img, im0s)

        # Process detections
        for i, det in enumerate(pred):  # detections per image
            if webcam:  # batch_size >= 1
                p, s, im0, frame = path[i], '%g: ' % i, im0s[i].copy(), dataset.count
            else:
                p, s, im0, frame = path, '', im0s, getattr(dataset, 'frame', 0)

            p = Path(p)  # to Path
            save_path = str(save_dir / p.name)  # img.jpg
            txt_path = str(save_dir / 'labels' / p.stem) + ('' if dataset.mode == 'image' else f'_{frame}')  # img.txt
            s += '%gx%g ' % img.shape[2:]  # print string
            gn = torch.tensor(im0.shape)[[1, 0, 1, 0]]  # normalization gain whwh
            if len(det):
                # Rescale boxes from img_size to im0 size
                det[:, :4] = scale_coords(img.shape[2:], det[:, :4], im0.shape).round()

                # Print results
                for c in det[:, -1].unique():
                    n = (det[:, -1] == c).sum()  # detections per class
                    s += f"{n} {names[int(c)]}{'s' * (n > 1)}, "  # add to string

                # Write results
                for *xyxy, conf, cls in reversed(det):
                    if save_txt:  # Write to file
                        xywh = (xyxy2xywh(torch.tensor(xyxy).view(1, 4)) / gn).view(-1).tolist()  # normalized xywh
                        line = (cls, *xywh, conf) if opt.save_conf else (cls, *xywh)  # label format
                        with open(txt_path + '.txt', 'a') as f:
                            f.write(('%g ' * len(line)).rstrip() % line + '\n')
                            
                    # write jiang #################
                    if save_json:
                        name = os.path.split(txt_path)[-1]
#                         print(name)

                        x1, y1, x2, y2 = float(xyxy[0]), float(xyxy[1]), float(xyxy[2]), float(xyxy[3])
                        bbox = [x1, y1, x2, y2]
                        img_name = name
                        conf = float(conf)

                        #add solution remove other
                        result.append(
                            {'name': img_name+'.jpg', 'category': int(cls+1), 'bbox': bbox,
                             'score': conf})
                        print("result: ", {'name': img_name+'.jpg', 'category': int(cls+1), 'bbox': bbox,'score': conf})


                    if save_img or view_img:  # Add bbox to image
                        label = f'{names[int(cls)]} {conf:.2f}'
                        plot_one_box(xyxy, im0, label=label, color=colors[int(cls)], line_thickness=3)

            # Print time (inference + NMS)
            print(f'{s}Done. ({t2 - t1:.3f}s)')

            # Stream results
            if view_img:
                cv2.imshow(str(p), im0)
                cv2.waitKey(1)  # 1 millisecond

            # Save results (image with detections)
            if save_img:
                if dataset.mode == 'image':
                    cv2.imwrite(save_path, im0)
                else:  # 'video'
                    if vid_path != save_path:  # new video
                        vid_path = save_path
                        if isinstance(vid_writer, cv2.VideoWriter):
                            vid_writer.release()  # release previous video writer

                        fourcc = 'mp4v'  # output video codec
                        fps = vid_cap.get(cv2.CAP_PROP_FPS)
                        w = int(vid_cap.get(cv2.CAP_PROP_FRAME_WIDTH))
                        h = int(vid_cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
                        vid_writer = cv2.VideoWriter(save_path, cv2.VideoWriter_fourcc(*fourcc), fps, (w, h))
                    vid_writer.write(im0)

    if save_txt or save_img:
        s = f"\n{len(list(save_dir.glob('labels/*.txt')))} labels saved to {save_dir / 'labels'}" if save_txt else ''
        print(f"Results saved to {save_dir}{s}")
    
    if save_json:
        with open(os.path.join("/result.json"), 'w') as fp:
            json.dump(result, fp, indent=4, ensure_ascii=False)

    print(f'Done. ({time.time() - t0:.3f}s)')


if __name__ == '__main__':
    parser = argparse.ArgumentParser()
    parser.add_argument('--weights', nargs='+', type=str, default='./runs/train/exp3/weights/best.pt', help='model.pt path(s)')
    parser.add_argument('--source', type=str, default='/defect', help='source')  # file/folder, 0 for webcam
    parser.add_argument('--save_dir', type=str, default='/', help='result save dir')
    parser.add_argument('--img-size', type=int, default=640, help='inference size (pixels)')
    parser.add_argument('--conf-thres', type=float, default=0.25, help='object confidence threshold')
    parser.add_argument('--iou-thres', type=float, default=0.45, help='IOU threshold for NMS')
    parser.add_argument('--device', default='', help='cuda device, i.e. 0 or 0,1,2,3 or cpu')
    parser.add_argument('--view-img', action='store_true', help='display results')
    parser.add_argument('--save-txt', action='store_true', help='save results to *.txt')
    parser.add_argument('--save-conf', action='store_true', help='save confidences in --save-txt labels')
    parser.add_argument('--classes', nargs='+', type=int, help='filter by class: --class 0, or --class 0 2 3')
    parser.add_argument('--agnostic-nms', action='store_true', help='class-agnostic NMS')
    parser.add_argument('--augment', action='store_true', help='augmented inference')
    parser.add_argument('--update', action='store_true', help='update all models')
    parser.add_argument('--project', default='runs/detect', help='save results to project/name')
    parser.add_argument('--name', default='exp', help='save results to project/name')
    parser.add_argument('--exist-ok', action='store_true', help='existing project/name ok, do not increment')
    opt = parser.parse_args()
    print(opt)
    check_requirements()

    with torch.no_grad():
        if opt.update:  # update all models (to fix SourceChangeWarning)
            for opt.weights in ['yolov5s.pt', 'yolov5m.pt', 'yolov5l.pt', 'yolov5x.pt']:
                detect()
                strip_optimizer(opt.weights)
        else:
            detect()

build镜像
docker build -t registry.cn-shenzhen.aliyuncs.com/your namespace:cv1.2 .
push镜像
docker push registry.cn-shenzhen.aliyuncs.com/your namespace:cv1.2
最终结果