Image recovery method and coding end

Abstract

The embodiment of the invention discloses an image recovery method and device, and the method comprises the steps: obtaining a reference candidate frame according to the inter-frame similarity and thequality of a reconstructed image when determining that the image quality needs to be adjusted; and allocating bits to a target frame according to the reference candidate frame, and performing video coding on the target frame based on the bits allocated to the target frame. Therefore, the reference candidate frame is adaptively adjusted through the inter-frame similarity and the reconstructed image quality, so that the bit allocation of the target frame is adjusted, the network packet loss of the image frame is reduced, the packet loss resistance of the image is effectively improved, and the better image quality and fluency are obtained while the image is quickly recovered at the decoding end.

Description

technical field [0001] Embodiments of the present invention relate to but are not limited to video encoding and decoding technologies, and more specifically relate to an image restoration method and an encoding end. Background technique [0002] Image restoration is the key technology of real-time video codec. In the transmission process under the constraint of limited bandwidth, because the transmission status of the network is dynamically changing, bandwidth overflow is easy to occur, resulting in data loss. [0003] figure 1 It is a schematic structural diagram of a video codec system in the related art. Such as figure 1 As shown, the video codec system includes an encoding end and a decoding end, the encoding end includes a video encoding device, and the decoding end includes a video decoding device. The video encoding device is used to encode the input video data, the encoded video sequence is transmitted to the decoding end through the video transmission channel, a...

AI Technical Summary

Problems solved by technology

Under the source, channel, and sink image restoration mechanism, when the image still cannot be restored, the technique adopted is to apply for an airspace I frame. The so-called airspace I frame means that the previous image frame is not referred to when encoding and decoding, so as to avoid even if the previous image frame If the image frame is lost, it will not lead to the loss of the current image frame; however, the image codes of the airspace I frame are relatively large. In the harsh network environment with limited bandwidth, the larger the image code, it will naturally cause a vicious cycle of network transmission and accelerate The loss of image frames under limited bandwidth makes it impossible to guarantee ideal image quality

[0004] In related technologies, there is no reasonable image restoration mechanism, and it is difficult to guarantee ideal image quality

Images