Design Of Backlight Image Detection And Correction System Based On FPGA

With the development of science and technology,digital imaging equipment is more and more widely used in our everyday life.However,in actual use,due to the influence of lighting and variations in shooting angles,backlight images will inevitably occur.Backlight will cause a decline in the quality of the image,such as blur and darkness.Therefore,it is necessary to deal with backlight images,in order to obtain better visual effect.However,in the real-time video image processing it has a high demand on the processing speed and data transmission rate.Software processing is difficult to meet these requirements.In recent years,with the development of semiconductor technology,the advantages of digital hardware system based on FPGA are more and more obvious in the field of real-time image processing.As a field programmable logic device FPGA,the amount of data throughput is very large,and in parallel operation,speed and performance FPGA is better than the DSP and ARM.Now in the high-end FPGA chip FPGA has integrated Nios II core,DSP module and the hardware multiplier,which can complete the complex operation.FPGA is the favorable choice of real-time image processing system.The embedded image processing system based on FPGA has the advantages of small volume,low power consumption,fast speed and many other advantages,which has been widely used in the field of image processing,and is the development trend of high performance embedded system in the future.In this paper,with the research performed on many backlight image correction algorithms and correction systems,it has been found that most existing backlight image correction systems only process the input images without evaluating them,and do not distinguish between backlight images and normal illumination image s.Therefore,this paper is based on the platform of FPGA to design an adaptive backlight image correction system.The main research works in this paper are as follows: Firstly,this paper presents a new method of backlight image detection based on edge details.By analyzing the change of the details of foreground image before and after the Gamma transform,we can determine the occurrence of backlight.Second,the traditional multi-scale correction algorithm will appear the halo phenomenon.This paper proposes an improved correction algorithm based on fusion of multi-scale Retinex.Multi-scale Retinex algorithm is performed on the background and the foreground of the original image.The outputs are fused with the original image.Third,the multi-scale Retinex algorithm is difficult to realize in hardware,this paper uses the method of look-up table and pipelining to solve the difficult problem of complex operation in FPGA.The experimental results show that the processing effect of the system is similar to the MATLAB simulation results.