Infrared Image Detail Enhancement Based On FPGA

With the rapid development of infrared thermography technology, infrared images have been widely used in military, medical, business and daily life. Due to the limit of imaging system and the disturbance of surrounding environment, the resolution and contrast of infrared images were usually low, unable to highlight the detail information of image texture. Therefore, the process of detail enhancement for infrared images should be arranged before reading or analyzing the image. The technology of detail enhancement for infrared images is also being studied more and more widely.The general characteristics of infrared images and the basic concepts of image enhancement were introduced firstly. The technique of digital detail enhancement (DDE) and the classical methods of image enhancement were also described in this thesis. On this basis, an optimized method of image enhancement on account of image weight was proposed. In this algorithm, the detail and background components of the image are separated by low-pass filtering firstly, and the initial image weight of each pixel is set as the difference of expansion filtering and erosion filtering. The image weight of the pixel points is further processed to deal with the problem of too dark or too bright pixels. Then the final image weight of each pixel is obtained. Finally, the detail image is handled by histogram equalization based on image weight, and the background of the image is further compressed by power transformation.The software simulation was carried out based on MATLAB and C platform. Then the algorithm was implemented with Verilog hardware language, which was implemented on the FPGA chip EP4CE115F29C7 of Altera. In the process of hardware implementation, a complete image processing system is built by the control of AD converter chip and the interface control of VGA. In the process of hardware implementation, a complete image processing system was built by the control of AD converter chip and the interface control of VGA. In the final hardware simulation and download test, the results of image processing were consistent with the results of software simulation. By the algorithm, the contrast and resolution of the infrared image were obviously improved. The image processing result was displayed on the LCD screen stably in the final, and the real-time performance has been implemented.