將COCO格式數據轉爲VOC格式

參考：

https://blog.csdn.net/weixin_39881922/article/details/85120379

將COCO格式數據轉爲VOC格式

'''
把coco數據集合的所有標註轉換到voc格式，不改變圖片命名方式，
注意，原來有一些圖片是黑白照片，檢測出不是 RGB 圖像，這樣的圖像不會被放到新的文件夾中
更新日期：2019-11-19
'''
from pycocotools.coco import \
    COCO  # 這個包可以從git上下載https://github.com/cocodataset/cocoapi/tree/master/PythonAPI，也可以直接用修改後的coco.py
import os, cv2, shutil
from lxml import etree, objectify
from tqdm import tqdm
from PIL import Image

CKimg_dir = '/media/zlee/Elements/mmdetection/coco2VOC/images'
CKanno_dir = '/media/zlee/Elements/mmdetection/coco2VOC/annotations'


# 若模型保存文件夾不存在，創建模型保存文件夾，若存在，刪除重建
def mkr(path):
    if os.path.exists(path):
        shutil.rmtree(path)
        os.mkdir(path)
    else:
        os.mkdir(path)


def save_annotations(filename, objs, filepath):
    annopath = CKanno_dir + "/" + filename[:-3] + "xml"  # 生成的xml文件保存路徑
    dst_path = CKimg_dir + "/" + filename
    img_path = filepath
    img = cv2.imread(img_path)
    im = Image.open(img_path)
    if im.mode != "RGB":
        print(filename + " not a RGB image")
        im.close()
        return
    im.close()
    shutil.copy(img_path, dst_path)  # 把原始圖像複製到目標文件夾
    E = objectify.ElementMaker(annotate=False)
    anno_tree = E.annotation(
        E.folder('1'),
        E.filename(filename),
        E.source(
            E.database('CKdemo'),
            E.annotation('VOC'),
            E.image('CK')
        ),
        E.size(
            E.width(img.shape[1]),
            E.height(img.shape[0]),
            E.depth(img.shape[2])
        ),
        E.segmented(0)
    )
    for obj in objs:
        E2 = objectify.ElementMaker(annotate=False)
        anno_tree2 = E2.object(
            E.name(obj[0]),
            E.pose(),
            E.truncated("0"),
            E.difficult(0),
            E.bndbox(
                E.xmin(obj[2]),
                E.ymin(obj[3]),
                E.xmax(obj[4]),
                E.ymax(obj[5])
            )
        )
        anno_tree.append(anno_tree2)
    etree.ElementTree(anno_tree).write(annopath, pretty_print=True)


def showbycv(coco, dataType, img, classes, origin_image_dir, verbose=False):
    filename = img['file_name']
    filepath = os.path.join(origin_image_dir, dataType, filename)
    I = cv2.imread(filepath)
    annIds = coco.getAnnIds(imgIds=img['id'], iscrowd=None)
    anns = coco.loadAnns(annIds)
    objs = []
    for ann in anns:
        name = classes[ann['category_id']]
        if 'bbox' in ann:
            bbox = ann['bbox']
            xmin = (int)(bbox[0])
            ymin = (int)(bbox[1])
            xmax = (int)(bbox[2] + bbox[0])
            ymax = (int)(bbox[3] + bbox[1])
            obj = [name, 1.0, xmin, ymin, xmax, ymax]
            objs.append(obj)
            if verbose:
                cv2.rectangle(I, (xmin, ymin), (xmax, ymax), (255, 0, 0))
                cv2.putText(I, name, (xmin, ymin), 3, 1, (0, 0, 255))
    save_annotations(filename, objs, filepath)
    if verbose:
        cv2.imshow("img", I)
        cv2.waitKey(0)


def catid2name(coco):  # 將名字和id號建立一個字典
    classes = dict()
    for cat in coco.dataset['categories']:
        classes[cat['id']] = cat['name']
        # print(str(cat['id'])+":"+cat['name'])
    return classes


def get_CK5(origin_anno_dir, origin_image_dir, verbose=False):
    dataTypes = ['train2017', 'val2017']
    for dataType in dataTypes:
        annFile = 'instances_{}.json'.format(dataType)
        annpath = os.path.join(origin_anno_dir, annFile)
        coco = COCO(annpath)
        classes = catid2name(coco)
        imgIds = coco.getImgIds()
        # imgIds=imgIds[0:1000]#測試用，抽取10張圖片，看下存儲效果
        for imgId in tqdm(imgIds):
            img = coco.loadImgs(imgId)[0]
            showbycv(coco, dataType, img, classes, origin_image_dir, verbose=False)


def main():
    base_dir = '/media/zlee/Elements/mmdetection/coco2VOC'  # step1 這裏是一個新的文件夾，存放轉換後的圖片和標註
    image_dir = os.path.join(base_dir, 'images')  # 在上述文件夾中生成images，annotations兩個子文件夾
    anno_dir = os.path.join(base_dir, 'annotations')
    mkr(image_dir)
    mkr(anno_dir)
    origin_image_dir = '/media/zlee/Elements/mmdetection/data/coco/images'  # step 2原始的coco的圖像存放位置
    origin_anno_dir = '/media/zlee/Elements/mmdetection/data/coco/annotations'  # step 3 原始的coco的標註存放位置
    verbose = True  # 是否需要看下標記是否正確的開關標記，若是true,就會把標記展示到圖片上
    get_CK5(origin_anno_dir, origin_image_dir, verbose)


if __name__ == "__main__":
    main()