Research On Optimization And Preprocessing Algorithm Of Bayer CCD Front - End Video Capture Module

The Charge Coupled Device (CCD) has great advantage of wide dynamic range, high-sensitivity, high Signal to Noise Ratio (SNR), high-level charge-transfer efficiency, and so on. Therefore, the CCD imaging sensor occupies an important position in the field of optoelectronic imaging. The video capture system based on area array CCD has achieved rapid development, which has developed from analog to digital, from standard definition to high definition rapidly and has huge develop potential and wide application foreground. However, the core technology has been kept secret strictly by foreign manufacturers for a long time. For this reason, there is important implication to research on video collecting system, especially on front-end video capture module and pretreatment algorithm independently.In this dissertation, color planar array CCD with CFA of Bayer has been taken as research object. Current situation and development tendency of CCD image sensor and video capture system are presented, the theories of CCD imaging device and Bayer CFA are analyzed, and design scheme of front-end video capture module in high-definition area CCD with Bayer CFA, which is closely linking with demand, are put forward in this paper. This design mainly includes two parts, one is driving schedule generator and configuration controller based on FPGA (Field Programmable Gate Array), the other part is hardware circuit containing driving circuit, analog-digital conversion circuit and power supply circuit. On this basis, the intense research and elaborate analysis of driving process of CCD are conducted combining with its chip architecture. With which as the theory base, optimum proposal for timing driven design and PCB board design are provided. In addition, for the characteristics of Bayer CFA and multi-channel output, an improved color-demosaicing algorithm and nonuniformity correction algorithm are also proposed. So far, the designing and optimizing of front-end video capture module for Bayer CCD are completed, and the algorithms of image pretreatment are studied.The debugging results show that the optimization design of front-end video capture module for color planar array Bayer CCD and algorithms of image pretreatment in this article have the characteristics of low noise, high-resolution and stable performance, successfully achieved the video output with format of 25fps and 1080p, which can fully satisfy the market.