Python cv2.selectROI() Examples

def select_roi(target):
    # 创建一个窗口
    cv2.namedWindow("image", flags= cv2.WINDOW_NORMAL | cv2.WINDOW_FREERATIO)
    cv2.imshow("image", target)
    # 是否显示网格 
    showCrosshair = True

    # 如果为Ture的话 , 则鼠标的其实位置就作为了roi的中心
    # False: 从左上角到右下角选中区域
    fromCenter = False
    # Select ROI
    rect = cv2.selectROI("image", target, showCrosshair, fromCenter)

    print("选中矩形区域")
    (x, y, w, h) = rect

    # Crop image
    roi = target[y : y+h, x:x+w]
    
    return rect, roi 

def write_to_trackfile(track_info):
    # st = time.time()
    with open(WORK_DIR + "/METADATA/" + "trackfile.txt", "a+") as f:
        for item in track_info:
            f.write(str(item) + "\n")
            # print(str(item) + "\n")
        f.close()
    # et = time.time()
    # print ("write time:", et - st)
    # sys.exit()
    
#uncomment to use as stand-alone file
# start_frame = 8*3
# vid = VIDEO_PATH + "/cam_8.avi"
# cap = cv2.VideoCapture(vid)
# cap.set(cv2.CAP_PROP_FPS, 8)
# cap.set(cv2.CAP_PROP_POS_FRAMES, start_frame)
# ok, frame = cap.read()
# bbox = selectROI(frame)
# track(0,cap,bbox,2)


# cam_id, cap, bbox, data_inc 

def youOnlyCropOnce(inputdir,outputdir):
    filesFound=[]
    ########### FIND FILES ###########
    for i in os.listdir(inputdir):
        if i.endswith(('.avi', '.mp4', '.mov', 'flv')):
            filesFound.append(os.path.join(inputdir,i))
    filenames=filesFound[0]
    #extract one frame
    currentDir = str(os.path.dirname(filenames))
    videoName = str(os.path.basename(filenames))
    os.chdir(currentDir)
    cap = cv2.VideoCapture(videoName)
    cap.set(1, 0)
    ret, frame = cap.read()
    fileName = str(0) + str('.bmp')
    filePath = os.path.join(currentDir, fileName)
    cv2.imwrite(filePath, frame)
    #find ROI

    img = cv2.imread(filePath)
    cv2.namedWindow('Select ROI', cv2.WINDOW_NORMAL)
    ROI = cv2.selectROI("Select ROI", img)
    width = abs(ROI[0] - (ROI[2] + ROI[0]))
    height = abs(ROI[2] - (ROI[3] + ROI[2]))
    topLeftX = ROI[0]
    topLeftY = ROI[1]
    cv2.waitKey(0)
    cv2.destroyAllWindows()

    for i in filesFound:
        #crop video with ffmpeg
        fileOut, fileType = i.split(".", 2)
        fileOutName = outputdir + '\\'+ os.path.basename(str(fileOut)) + str('_cropped.')+ str(fileType)
        command = str('ffmpeg -i ') +'"'+ str(i) +'"'+ str(' -vf ') + str('"crop=') + str(width) + ':' + str(
            height) + ':' + str(topLeftX) + ':' + str(topLeftY) + '" ' + str('-c:v libx264 -crf 21 -c:a copy ') +'"'+ str(
            fileOutName)+'"'
        total = width + height + topLeftX + topLeftY

        file = pathlib.Path(fileOutName)
        if file.exists():
            print(os.path.basename(fileOutName), 'already exist')
        else:
            if width==0 and height ==0:
                print('Video not cropped')
            elif total != 0:
                print('Cropping video...')
                print(command)
                subprocess.call(command, shell=True)
            elif total ==0:
                print('Video not cropped')

    os.remove(filePath)
    print('Process completed.') 

def main():
    # debug = 0 , no log will produce
    # debug = 1 , will produce log file
    tracker = SiameseTracker(debug=0)
    time_per_frame = 0

    if len(sys.argv) <= 1:
        print('[ERROR]: File path error!')
        return

    if sys.argv[1] == "cam":
        cap = cv2.VideoCapture(0)
    else:
        cap = cv2.VideoCapture(sys.argv[1])

    while True:
        # Capture frame-by-frame
        ret, frame = cap.read()
        frame = preprocess(frame)
        cv2.imshow('frame', postprocess(frame))
        if cv2.waitKey(1500) & 0xFF == ord('o'):
            break

    # select ROI and initialize the model
    r = cv2.selectROI(postprocess(frame))
    cv2.destroyWindow("ROI selector")
    print('ROI:', r)
    tracker.set_first_frame(frame, r)

    while True:
        ret, frame = cap.read()
        frame = preprocess(frame)
        start_time = datetime.datetime.now()
        reported_bbox = tracker.track(frame)
        end_time = datetime.datetime.now()

        # Display the resulting frame
        # print(reported_bbox)
        cv2.rectangle(frame, (int(reported_bbox[0]), int(reported_bbox[1])),
                      (
                          int(reported_bbox[0]) + int(reported_bbox[2]),
                          int(reported_bbox[1]) + int(reported_bbox[3])),
                      (0, 0, 255), 2)

        duration = end_time - start_time
        time_per_frame = 0.9 * time_per_frame + 0.1 * duration.microseconds
        cv2.putText(frame, 'FPS ' + str(round(1e6 / time_per_frame, 1)),
                    (30, 50), 0, 1, (0, 0, 255), 3)

        cv2.imshow('frame', postprocess(frame))

        if cv2.waitKey(1) & 0xFF == ord('q'):
            break

    # When everything done, release the capture
    cap.release()
    cv2.destroyAllWindows() 

def cropvid(filenames):
    if filenames:

        #extract one frame
        currentDir = str(os.path.dirname(filenames))
        videoName = str(os.path.basename(filenames))
        os.chdir(currentDir)
        cap = cv2.VideoCapture(videoName)
        cap.set(1, 0)
        ret, frame = cap.read()
        fileName = str(0) + str('.bmp')
        filePath = os.path.join(currentDir, fileName)
        cv2.imwrite(filePath, frame)
        #find ROI

        img = cv2.imread(filePath)
        cv2.namedWindow('Select ROI', cv2.WINDOW_NORMAL)
        ROI = cv2.selectROI("Select ROI", img)
        width = abs(ROI[0] - (ROI[2] + ROI[0]))
        height = abs(ROI[2] - (ROI[3] + ROI[2]))
        topLeftX = ROI[0]
        topLeftY = ROI[1]
        cv2.waitKey(0)
        cv2.destroyAllWindows()

        #crop video with ffmpeg
        fileOut, fileType = videoName.split(".", 2)
        fileOutName = str(fileOut) + str('_cropped.mp4')
        command = str('ffmpeg -i ') +'"'+ str(videoName) +'"'+ str(' -vf ') + str('"crop=') + str(width) + ':' + str(
            height) + ':' + str(topLeftX) + ':' + str(topLeftY) + '" ' + str('-c:v libx264 -crf 21 -c:a copy ') +'"'+ str(
            fileOutName)+'"'
        total = width + height + topLeftX + topLeftY

        file = pathlib.Path(fileOutName)
        if file.exists():
            print(os.path.basename(fileOutName), 'already exist')
        else:
            if width==0 and height ==0:
                print('Video not cropped')
            elif total != 0:
                print('Cropping video...')
                print(command)
                subprocess.call(command, shell=True)
                os.remove(filePath)
                print('Cropped video saved!')
                return fileOutName
            elif total ==0:
                print('Video not cropped')

        os.remove(filePath)

    else:
        print('Please select a video to crop') 

def cropvid_auto(filenames,outputdir):
    global width,height

    #extract one frame
    currentDir = str(os.path.dirname(filenames))
    videoName = str(os.path.basename(filenames))
    os.chdir(currentDir)
    cap = cv2.VideoCapture(videoName)
    cap.set(1, 0)
    ret, frame = cap.read()
    fileName = str(0) + str('.bmp')
    filePath = os.path.join(currentDir, fileName)
    cv2.imwrite(filePath, frame)

    #find ROI

    img = cv2.imread(filePath)
    cv2.namedWindow('Select ROI', cv2.WINDOW_NORMAL)
    ROI = cv2.selectROI("Select ROI", img)
    width = abs(ROI[0] - (ROI[2] + ROI[0]))
    height = abs(ROI[2] - (ROI[3] + ROI[2]))
    topLeftX = ROI[0]
    topLeftY = ROI[1]
    cv2.waitKey(0)
    cv2.destroyAllWindows()

    #crop video with ffmpeg
    fileOut, fileType = videoName.split(".", 2)
    fileOutName = str(fileOut) + str('_cropped.mp4')



    total = width+height+topLeftX +topLeftY

    if total != 0:
        command = (str('ffmpeg -y -i ') + str(outputdir) + '\\' + str(videoName) + str(' -vf ') + str('"crop=') + str(width) + ':' + str(height) + ':' + str(topLeftX) + ':' + str(topLeftY) + '" ' + str('-c:v libx264 -c:a copy ') + str(os.path.join(outputdir, fileOutName)) + '\n'
            'move \"' + str(outputdir) + '\\' + videoName + '\" \"' + (outputdir) + '\\' + 'tmp\"' + '\n'
             'copy \"' + str(outputdir) + '\\' + os.path.basename(fileOutName) + '\" \"' + (outputdir) + '\\' + 'tmp\"' + '\n'
            'rename \"' + os.path.join(str(outputdir), os.path.basename(fileOutName)) + '\" \"' + os.path.basename(videoName) + '\"')
        print(videoName,'added into the crop video queue.')
        os.remove(filePath)
    elif total == 0:
        command = []
        print('nothing added to the script as no coordinates was selected')

    if os.path.exists(filePath):
        os.remove(filePath)
    return command 

def cropvid_queue(filenames,outputdir):
    global width,height
    #extract one frame
    currentDir = str(os.path.dirname(filenames))
    videoName = str(os.path.basename(filenames))
    os.chdir(currentDir)
    cap = cv2.VideoCapture(videoName)
    cap.set(1, 0)
    ret, frame = cap.read()
    fileName = str(0) + str('.bmp')
    filePath = os.path.join(currentDir, fileName)
    cv2.imwrite(filePath, frame)

    #find ROI

    img = cv2.imread(filePath)
    cv2.namedWindow('Select ROI', cv2.WINDOW_NORMAL)
    ROI = cv2.selectROI("Select ROI", img)
    width = abs(ROI[0] - (ROI[2] + ROI[0]))
    height = abs(ROI[2] - (ROI[3] + ROI[2]))
    topLeftX = ROI[0]
    topLeftY = ROI[1]
    cv2.waitKey(0)
    cv2.destroyAllWindows()

    #crop video with ffmpeg
    fileOut, fileType = videoName.split(".", 2)
    fileOutName = str(fileOut) + str('_cropped.mp4')

    total = width+height+topLeftX +topLeftY

    if total != 0:
        command = (str('ffmpeg -y -i ')+ '"' + str(outputdir) + '\\' + str(videoName)+ '"' + str(' -vf ') + str('"crop=') + str(width) + ':' + str(height) + ':' + str(topLeftX) + ':' + str(topLeftY) + '" ' + str('-c:v libx264 -c:a copy ') + '"'+ str(os.path.join(outputdir, fileOutName))+ '"' + '\n'
            'move \"' + str(outputdir) + '\\' + videoName + '\" \"' + (outputdir) + '\\' + 'tmp\"' + '\n'
             'copy \"' + str(outputdir) + '\\' + os.path.basename(fileOutName) + '\" \"' + (outputdir) + '\\' + 'tmp\"' + '\n'
            'rename \"' + os.path.join(str(outputdir), os.path.basename(fileOutName)) + '\" \"' + os.path.basename(videoName) + '\"')
        print(videoName, 'added into the crop video queue.')
        os.remove(filePath)
        return command
    else:
        print('nothing added to the script as no coordinates was selected')
        pass
    if os.path.exists(filePath):
        os.remove(filePath) 

def main():
    # load config
    cfg.merge_from_file(args.config)
    cfg.CUDA = torch.cuda.is_available() and cfg.CUDA
    device = torch.device('cuda' if cfg.CUDA else 'cpu')

    # create model
    model = ModelBuilder()

    # load model
    model.load_state_dict(torch.load(args.snapshot,
        map_location=lambda storage, loc: storage.cpu()))
    model.eval().to(device)

    # build tracker
    tracker = build_tracker(model)

    first_frame = True
    if args.video_name:
        video_name = args.video_name.split('/')[-1].split('.')[0]
    else:
        video_name = 'webcam'
    cv2.namedWindow(video_name, cv2.WND_PROP_FULLSCREEN)
    for frame in get_frames(args.video_name):
        if first_frame:
            try:
                init_rect = cv2.selectROI(video_name, frame, False, False)
            except:
                exit()
            tracker.init(frame, init_rect)
            first_frame = False
        else:
            outputs = tracker.track(frame)
            if 'polygon' in outputs:
                polygon = np.array(outputs['polygon']).astype(np.int32)
                cv2.polylines(frame, [polygon.reshape((-1, 1, 2))],
                              True, (0, 255, 0), 3)
                mask = ((outputs['mask'] > cfg.TRACK.MASK_THERSHOLD) * 255)
                mask = mask.astype(np.uint8)
                mask = np.stack([mask, mask*255, mask]).transpose(1, 2, 0)
                frame = cv2.addWeighted(frame, 0.77, mask, 0.23, -1)
            else:
                bbox = list(map(int, outputs['bbox']))
                cv2.rectangle(frame, (bbox[0], bbox[1]),
                              (bbox[0]+bbox[2], bbox[1]+bbox[3]),
                              (0, 255, 0), 3)
            cv2.imshow(video_name, frame)
            cv2.waitKey(40)