Image processing method, image processor, image processing module and equipment

Abstract

The embodiment of the invention provides an image processing method, an image processor, an image processing module and equipment, the image processing method is applied to the image processor, and the image processing method comprises the following steps: obtaining a to-be-processed image; a target output image corresponding to the to-be-processed image is obtained at least by querying a lookup table, the lookup table is used for storing output data corresponding to all traversal image blocks in a plurality of traversal image blocks, and the traversal image blocks are image blocks determined by a receptive field of a target neural network. Compared with an image processor scheme realized by adopting a neural network, the image processing method provided by the embodiment of the invention has the advantages that the running speed is greatly improved, and consumed logic and hardware resources are obviously reduced.

Description

technical field [0001] The present application relates to the field of image processing, and specifically, the embodiments of the present application relate to an image processing method, an image processor, an image processing module, and a device. Background technique [0002] Image Processor ISP (Image Signal Process) is a processing pipeline that converts raw images (eg, RAW data in Bayer mosaic format) output by image sensors (eg, CMOS sensors and CCD sensors) into RGB or YUV images . For example, for camera products, it is also necessary to further convert the RGB format or YUV image output by the image processor into JPEG format for convenient storage. [0003] With the rise of deep learning technology (Deep Learning), some people proposed to use neural network to replace the related technology image processor ISP and achieved good results. There is no advantage in area and cannot be widely used in practical scenarios. SUMMARY OF THE INVENTION [0004] The purpos...

AI Technical Summary

Problems solved by technology

[0003] With the rise of deep learning technology (Deep Learning), some people have proposed to use neural networks to replace the image processor ISP of related technologies and achieved good results. However, due to the complexity of the neural network model, such technical solutions are time and chip There is no advantage in area and cannot be widely used in actual scenarios

Images