基于侧抑制系数的红外图像细节增强算法

针对红外图像对比度低、细节不清晰和视觉效果模糊的缺点，提出了基于侧抑制系数的红外图像细节增强算法。通过分析指数分布侧抑制系数计算方法的不足，提出利用二次函数分布计算侧抑制系数的方法，提高算法增强红外图像细节的能力，利用红外图像本身的灰度信息自适应地调整侧抑制系数的参数，进而自适应确定侧抑制网络，对图像目标和背景进行不同程度的抑制，结合临界可偏差(just noticeable difference,JND)曲线中得到的人眼视觉分辨率特性，对图像进行能量恢复调整，使图像均值保持在人眼适宜观察的状态。仿真结果显示，与自适应指数函数侧抑制算法相比，算法处理后的红外图像对比度、信息熵以及信噪比都得到提高，图像细节更加清晰，视觉效果得到明显改善。

Infrared imagedetail enhancement algorithm based on lateral-inhibition coefficient

Aiming at those disadvantages of infrared image such as low contrast, blurred details and visual effect, an infrared image detail enhancement algorithm based on lateral inhibition coefficient is proposed. Firstly, by analyzing the shortcomings of lateral inhibition coefficient of exponential distribution, the algorithm makes a quadratic function to calculate the coefficient and improves the ability of the algorithm to enhance the image details, and then adaptively adjusts the parameters of coefficient by the grayscale information of the infrared image itself. Then the lateral inhibition network is adaptively determined, and the target and background of the infrared image are suppressed in varying degrees. Finally, the algorithm keeps the mean of image in a state of suitable observation by energy recovery and human visual resolution characteristics obtained from just noticeable difference (JND) curve. Compared with the lateral inhibition algorithm based on adaptive exponential function, the experimental results show that the contrast, entropy and signal-to-noise ratio of the infrared image processed by the proposed algorithm are improved. Eventually, the image details are clearer and the visual effect of the image is improved significantly.