Study On Infrared Image Detail Enhancement Algorithm And Its Implementation

Infrared imaging technology is the signal processing technology that connects the infrared radiation among the environment and displays it in two-dimensional image. Due to the advantage of resistant to electric-magnetic interference, easy to hide, long working hours and high reliability, the infrared imaging technology has been used in many fields including medical diagnostics, industrial detection, military target tracing and so on. However, because of the weakness of low contrast, the image's quality is poor and it is not conducive to the subsequent identification of the target or the track detection processing. Therefore, it is necessary to process the raw infrared image with image enhancement algorithm. Even though the goal of better contrast and signal to noise ratio is able to be achieved with traditional image enhancement algorithms, the drawbacks of losing large amounts of detail and poor adaptivity are still existing, which has advers effects on the application of the infrared imaging technology.Given the research status mentioned above, this dissertation proposed an adaptive detail enhancement algorithm for infrared image base on the infrared image's characters after comparing those algorithms existed. This new algorithm can be divided into four steps. Firstly, the input raw data will be separated into two layers. Then the basic layer, which contains the background in low frequency, will be enhanced using the adaptive histogram projection algorithm based on its own statistic feature for improving contrast and adaptivity to the environment. And at the same time the detail layer, which contains the texture message in high frequency, will be handled using the algorithm based on noise visibility function for giving raise to its texture and suppressing the noise. The final output image will be generated from the two handled parts.Comparing those experimental results of different algorithm processing different real infrared image data of different environments, it shows that the proposed algorithm has good environmental adaptivity and can process those data efficiently. What's more, its output images have better contrast and more detail, which means better image quality. On the other hand, the results of its subjective and objective evaluation criteria suggest that the new algorithm has better ability of High Dynamic Range compression, contrast improvement and details reservation, as well as environment adaptivity.