Method and device for analyzing video stream

Abstract

The invention discloses a method and device for analyzing a video stream and relates to the communication technology of multimedia. The method comprises the following steps of: at the same time that network packets are captured, judging whether the session of each captured network packet is a video stream or not, and displaying the information of video stream sessions; and when a request for playing the video stream sessions is received, reading the network packets of the video stream sessions, decoding and playing. The judgment for the types of the network packets is carried out at the same time that the network packets are captured, and the network packets are sorted and managed, so that when the quality of a video is needed to be checked, the video stream sessions can be directly played, users can intuitively know the quality of images, and the efficiency for positioning the quality problem of the video is increased.

Description

technical field [0001] The invention relates to multimedia communication technology, in particular to a video stream analysis method and device. Background technique [0002] At present, with the development of communication technology, video services have been widely used. However, during the development and maintenance of video services, video quality often fails to meet the requirements. The main causes of video quality failure are as follows Two reasons: 1. Network problems, excessive packet loss rate, jitter, and delay of video streams. 2. Software problems. It may be that the software failure of the receiver makes the video stream that meets the quality requirements not be displayed correctly, or it may be because the video package sent by the sender is defective, such as a field with an incorrect value. [0003] In order to locate the above-mentioned video quality problems, the current common practice is to capture network packets on the sending side, receiving side,...

AI Technical Summary

Problems solved by technology

[0002] At present, with the development of communication technology, video services have been widely used. However, during the development and maintenance of video services, video quality often fails to meet the requirements. The main causes of video quality failure are as follows Two reasons: 1. Network problems, excessive packet loss rate, jitter, and delay of video streams

2. Software problems, it may be that the software failure of the receiver makes the video stream that meets the quality requirements not be presented correctly, or it may be because the video package sent by the sender is defective, such as a field value is wrong

[0004] 1. Due to the large amount of captured packet data, it is very difficult to locate a specific video call stream from the massive captured packet data and track it at the data packet level. First, the call stream is interleaved with other network packets Together, there may be a large number of network packets that are not related to the call

Even in the code stream of a call, it is very difficult to locate what error occurred at what time

[0005] 2. This kind of analysis is completely at the data level, and it is analyzed manually. The positioning efficiency is very low, and the positioning personnel need to have solid video codec knowledge and skills

[0006] 3. Analyzing the packet capture data is an after-the-fact analysis. It is often necessary to stop the packet capture and then open the packet capture file, which interrupts the capture of other ongoing video call data. This damages the integrity of the packet capture data. Most likely fatal, because a very important and difficult-to-reproduce error may be happening, and the packet capture has been interrupted

[0007] 4. Lack of intuition. When analyzing the captured packet data, it is impossible to correspond to the human brain impression of the image quality on site. For example, to what extent is the image stuck? To what extent is the image spent? To what extent? It should be said that the current analysis method is still at a relatively macro level, and the conclusion of the analysis is often the possibility of causing problems rather than a certainty

[0009] The above analysis shows that the currently used video stream analysis methods have large limitations and low efficiency in the process of quality positioning

Images