opencv 读取错误:[h264 @ 0x8f915e0] 解码 MB 53 20，字节流 -7 时出错 [英] opencv read error:[h264 @ 0x8f915e0] error while decoding MB 53 20, bytestream -7

问题描述

我的配置:

  ubuntu 16.04
  opencv 3.3.1
  gcc version 5.4.0 20160609
  ffmpeg version 3.4.2-1~16.04.york0

我用以下方法构建了 opencv:

and I built opencv with:

cmake -D CMAKE_BUILD_TYPE=RELEASE -D CMAKE_INSTALL_PREFIX=/usr/local -D PYTHON_EXECUTABLE=$(which python) -D OPENCV_EXTRA_MODULES_PATH=/home/xxx/opencv_contrib/modules -D WITH_QT=ON -D WITH_OPENGL=ON -D WITH_IPP=ON -D WITH_OPENNI2=ON -D WITH_V4L=ON -D WITH_FFMPEG=ON -D WITH_GSTREAMER=OFF -D WITH_OPENMP=ON -D WITH_VTK=ON -D BUILD_opencv_java=OFF -D BUILD_opencv_python3=OFF -D WITH_CUDA=ON -D ENABLE_FAST_MATH=1 -D WITH_NVCUVID=ON -D CUDA_FAST_MATH=ON -D BUILD_opencv_cnn_3dobj=OFF -D FORCE_VTK=ON  -D WITH_CUBLAS=ON -D CUDA_NVCC_FLAGS="-D_FORCE_INLINES" -D WITH_GDAL=ON -D WITH_XINE=ON -D BUILD_EXAMPLES=OFF -D BUILD_DOCS=ON -D BUILD_PERF_TESTS=OFF -D BUILD_TESTS=OFF  -D BUILD_opencv_dnn=OFF -D BUILD_PROTOBUF=OFF -D opencv_dnn_BUILD_TORCH_IMPORTER=OFF -D opencv_dnn_PERF_CAFFE=OFF -D opencv_dnn_PERF_CLCAFFE=OFF -DBUILD_opencv_dnn_modern=OFF -D CUDA_ARCH_BIN=6.1 ..

并使用这些python代码读取和显示:

and use these python code to read and show:

import cv2
from com.xxx.cv.core.Image import Image

capture=cv2.VideoCapture("rtsp://192.168.10.184:554/mpeg4?username=xxx&password=yyy")
while True:
    grabbed,content=capture.read()
    if grabbed:
        Image(content).show()
        doSomething()
    else:
        print "nothing grabbed.."

每次读取大约 50 帧后，都会出现如下错误:

Everytime, after reading about 50 frames,it will give an error like:

[h264 @ 0x8f915e0] error while decoding MB 53 20, bytestream -7

然后什么也抓不住了，奇怪的是:

then nothing can be grabbed further,and the strange thing is:

1,comment doSomething() or
2,keep doSomething() and recording the stream from same IPCamera,then run
  code against recorded video

两种情况，代码都运行良好，谁能告诉如何解决这个问题?提前致谢！

both cases,code works fine,can anyone tell how to solve this problem?Thank in advance!

推荐答案

我们先来看一个简单的RTSP阅读示例程序

Let's first look at a simple sample program for reading RTSP

import cv2
cap=cv2.VideoCapture("rtsp://admin:admin_123@172.0.0.0")

ret,frame = cap.read()
while ret:
    ret,frame = cap.read()
    cv2.imshow("frame",frame)
    if cv2.waitKey(1) & 0xFF == ord('q'):
        break
cv2.destroyAllWindows()
cap.release()

帧值是每一帧的图像.但是如果在while代码块中添加对每一帧的识别操作，比如添加Tensorflow来识别其中的小动物，就会报这样的错误，while循环会因为这个异常而中断.

The frame value is the image of each frame. But if you add the recognition operation for each frame in the while code block, such as adding Tensorflow to recognize the small animals in it, such an error will be reported and the while loop will be interrupted due to this exception.

[h264 @ 0x55abeda05080] left block unavailable for requested intra mode
[h264 @ 0x55abeda05080] error while decoding MB 0 14, bytestream 104435

原来FFMPEG Lib不支持rtsp协议中的H264视频，所以解决办法是写两个不同的线程分别处理每一帧的图像，然后另一个线程来处理每一帧的图像.

It turns out that FFMPEG Lib does not support H264 videos in the rtsp protocol, so the solution is to write two different threads to process the images of each frame separately, and then another thread to process the images of each frame.

思路如下:使用队列，采用先进先出策略，在一个线程开始接收数据，在另一个线程逐帧处理数据

The idea is as follows: use a queue, adopt a first-in-first-out strategy, start receiving data in one thread, and process frame-by-frame data in another thread

解决方案代码如下:

import cv2
import queue
import time
import threading
q=queue.Queue()

def Receive():
    print("start Reveive")
    cap = cv2.VideoCapture("rtsp://admin:admin_123@172.0.0.0")
    ret, frame = cap.read()
    q.put(frame)
    while ret:
        ret, frame = cap.read()
        q.put(frame)


def Display():
     print("Start Displaying")
     while True:
         if q.empty() !=True:
            frame=q.get()
            cv2.imshow("frame1", frame)
         if cv2.waitKey(1) & 0xFF == ord('q'):
                break
            
if __name__=='__main__':
    p1=threading.Thread(target=Receive)
    p2 = threading.Thread(target=Display)
    p1.start()
    p2.start()

Receive作为接收数据的线程，Display作为一个简单的进程进行展示.

Receive is used as a thread for receiving data, and Display is displayed as a simple process.

这篇关于opencv 读取错误:[h264 @ 0x8f915e0] 解码 MB 53 20，字节流 -7 时出错的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！