多段线性拉伸增强算法及其FPGA实现

Research on Multi-segment Linear Stretch Algorithm and Its FPGA Implementation

为了实现红外图像增强,以FPGA为平台,提出了一种新型的红外图像增强算法。在设计过程中,根据红外图像处理的特点,利用一帧图像的均值和标准差快速的将图像灰度范围划分成若干段,根据每段的拉伸系数做不同的线性灰度变换实现图像中目标对比度的增强。该算法不仅能实现自适应图像增强,也可以通过在线调节来特别增强某一段或某几段温度范围内的目标,达到目标识别的目的。经验证该算法优于一般的线性增强算法,获得了很好的视觉效果。

A new type of image enhancement algorithm on FPGA platform for infrared image is introduced in this paper. In the design, according to the characteristics of the infrared images, the gray-scale scope of an image is divided fast into several segments by the mean and standard deviation of the image, and then the algorithm applies different linear grayscale transformation depending on each stretch factor to achieve image contrast enhancement. This algorithm can realize adaptive image enhancement and also can enhance specially one or a few segments to identify the target. It has been improved that this algorithm is superior to usual linear enhancement algorithm and obtains a good visual effect.