ValueError: not enough values to unpack (expected 3, got 2)

[momo1986]

您好，感谢您的项目分享，我想做一个其他物体检测追踪的模型。 我有了一个预训练的图像识别分类模型。 然后我load视频做图像腐蚀的时候报错了。 这是我在windows下运行的error-log：

D:\PythonSpace\Hand_Tracking>python track_finger.py --file car.flv Using TensorFlow backend. car.flv D:\PythonSpace\Hand_Tracking D:\PythonSpace\Hand_Tracking\car.flv <class 'cv2.VideoCapture'> 2019-03-22 11:51:22.024142: I tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 C:\Python\Python37\lib\site-packages\keras\engine\saving.py:292: UserWarning: No training configuration found in save file: the model was not compiled. Compile it manually. warnings.warn('No training configuration found in save file: ' [[0 0 0 ... 0 0 0] [0 0 0 ... 0 0 0] [0 0 0 ... 0 0 0] ... [0 0 0 ... 0 0 0] [0 0 0 ... 0 0 0] [0 0 0 ... 0 0 0]] Traceback (most recent call last): File "track_finger.py", line 157, in if name == 'main': File "track_finger.py", line 44, in main args = parser.parse_args() File "track_finger.py", line 109, in track dilated = cv2.dilate(th, cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (8, 3)), iterations=2) ValueError: not enough values to unpack (expected 3, got 2)

几乎没有怎么改变您的原始代码，主要是换了个分类模型：

Date : 7th Jan 2018

Author : xiaochus
Email : xiaochus@live.cn
Affiliation : School of Computer Science and Communication Engineering
                - Jiangsu University - China

License : MIT

Status : Under Active Development

Description :
OpenCV 3 & Keras implementation of the vehicle tracking.
"""

import sys
import copy
import argparse
import cv2
import numpy as np
from keras.models import load_model
import os

from utils.entity import Entity

def main(argv):
    parser = argparse.ArgumentParser()
    # Required arguments.
    parser.add_argument(
        "--file",
        help="Input video file.",
    )
    # Optional arguments.
    parser.add_argument(
        "--iou",
        default=0.2,
        help="threshold for tracking",
    )
    args = parser.parse_args()
    track(args.file, args.iou)

def overlap(box1, box2):
    """
    Check the overlap of two boxes
    """
    endx = max(box1[0] + box1[2], box2[0] + box2[2])
    startx = min(box1[0], box2[0])
    width = box1[2] + box2[2] - (endx - startx)

    endy = max(box1[1] + box1[3], box2[1] + box2[3])
    starty = min(box1[1], box2[1])
    height = box1[3] + box2[3] - (endy - starty)

    if (width <= 0 or height <= 0):
        return 0
    else:
        Area = width * height
        Area1 = box1[2] * box1[3]
        Area2 = box2[2] * box2[3]
        ratio = Area / (Area1 + Area2 - Area)

        return ratio

def track(video, iou):
    print(video)
    print(os.getcwd())
    print(os.path.join(os.getcwd(), video))
    camera = cv2.VideoCapture(video)
    print(type(camera))
    res, frame = camera.read()
    y_size = frame.shape[0]
    x_size = frame.shape[1]

    # Load CNN classification model
    model = load_model('brightness3ch.h5')
    # Definition of MOG2 Background Subtraction
    bs = cv2.createBackgroundSubtractorMOG2(detectShadows=True)
    history = 20
    frames = 0
    counter = 0

    track_list = []
    cv2.namedWindow("detection", cv2.WINDOW_NORMAL)
    while True:
        res, frame = camera.read()
        y_size = frame.shape[0]
        x_size = frame.shape[1]
        cv2.imwrite("result.png", frame)
        if res is False:
            break
        # Train the MOG2 with first frames frame
        fg_mask = bs.apply(frame)

        if frames < history:
            frames += 1
            continue
        # Expansion and denoising the original frame
        th = cv2.threshold(fg_mask.copy(), 244, 255, cv2.THRESH_BINARY)[1]
        print(th)
        th = cv2.erode(th, cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (3, 3)), iterations=2)
        dilated = cv2.dilate(th, cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (8, 3)), iterations=2)
        image, contours, hier = cv2.findContours(dilated, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)

        # Check the bouding boxs
        for c in contours:
            x, y, w, h = cv2.boundingRect(c)
            if cv2.contourArea(c) > 3000:
                # Extract roi
                img = frame[y: y + h, x: x + w, :]
                rimg = cv2.resize(img, (64, 64), interpolation=cv2.INTER_CUBIC)
                image_data = np.array(rimg, dtype='float32')
                image_data /= 255.
                roi = np.expand_dims(image_data, axis=0)
                flag = model.predict(roi)
                print(flag)

                if flag[0][0] >= 0 and flag[0][0] <=9:
                    e = Entity(counter, (x, y, w, h), frame)

                    # Exclude existing targets in the tracking list
                    if track_list:
                        count = 0
                        num = len(track_list)
                        for p in track_list:
                            if overlap((x, y, w, h), p.windows) < iou:
                                count += 1
                        if count == num:
                            track_list.append(e)
                    else:
                        track_list.append(e)
                    counter += 1

        # Check and update goals
        if track_list:
            tlist = copy.copy(track_list)
            for e in tlist:
                x, y = e.center
                if 10 < x < x_size - 10 and 10 < y < y_size - 10:
                    e.update(frame)
                else:
                    track_list.remove(e)
        frames += 1
        cv2.imshow("detection", frame)
        if cv2.waitKey(110) & 0xff == 27:
            break
    camera.release()

if __name__ == '__main__':
    main(sys.argv)

在做膨胀腐蚀的时候出错了，用您的原始视频和我自己的视频都碰到了同样的error-log

[xiaochus]

@momo1986 是不是因为OpenCV版本的问题，这个代码是基于OpenCV 3.2的。

[momo1986]

@momo1986 是不是因为OpenCV版本的问题，这个代码是基于OpenCV 3.2的。

报错在cv2.dilate函数上 另外，我threshold一路拿到0数组是什么原因，也没有看到起来的图片，是一块黑色的区域。但我又可以存frame作为图片

[ShidiDaisy]

运行原代码我也遇到同样的问题了 请问有人解决了吗？