摘要
辐射场景下CMOS图像传感器受高能光子影响,成像图像存在对比度不足、偏色等视觉退化问题,严重干扰γ辐射场景的清晰化可视化效果。为提升场景视觉图像质量,提出一种基于多元重映射的γ辐射场景图像增强方法,从提升图像对比度和消除颜色失真2个方面实现γ辐射场景的清晰可视化效果。针对γ辐射场景图像对比度不足的问题,利用限制对比度直方图均衡方法在抑制图像结构信息过度增强的同时提升图像对比度。针对γ辐射场景图像偏色的问题,采用颜色直方图重映射方法改善各颜色分量动态响应不均衡。通过分段非线性重映射对亮、暗区域进行不同程度的校正,解决在对比度提升及颜色校正过程中因亮暗像素同等尺度调节引起的图像局部过调问题。在真实γ辐射场景图像上进行多组对比实验,结果表明,相比典型图像增强方法,该方法在颜色、对比度、信息熵量化指标及视觉呈现中均取得较优的效果,图像质量综合评价指标最高可为3.140 76。
Inγradiation scenarios,CMOS image sensors are susceptible to the influence of high-energy photons.This leads to visual degradation problems,including insufficient contrast and color distortion in the captured images,ultimately hindering clear visualization of radiation scenes.To improve the visual image quality of the scene,a method based on multi-variate remapping is proposed as aγradiation scene image enhancement method.It is implemented from two aspects:improving image contrast and eliminating color distortion to ensure clear visualization of radiation scenes.This method utilizes a limited contrast histogram equalization method to suppress excessive enhancement of image structural information while improvingγradiation scene image contrast.To address the issue of color deviation inγradiation scene images,a color histogram remapping method is used to improve the uneven dynamic response of each color component.This involves using segmented nonlinear remapping to correct bright and dark regions to varying degrees,thus resolving the problem of local image overshoot caused by adjusting the same scale of bright and dark pixels during contrast enhancement and color correction.In reality,multiple comparative experiments are conducted on radiation scene images.The experimental results indicate that,when compared to typical image enhancement methods,this method achieves superior results in terms of color,contrast,information entropy quantification indicators,and visual presentation.The highest comprehensive evaluation index for image quality can reach 3.14076.
作者
方琳琳
邓豪
张华
赵俊琴
FANG Linin;DENG Hao;ZHANG Hua;ZHAO Junqin(Robot Technology Used for Special Environment Key Laboratory of Sichuan Province,School of Information Engineering,Southwest University of Science and Technology,Mianyang 621010,Sichuan,China;Institute of Aerospace Technology,China Aerodynamics Research and Development Center,Mianyang 621000,Sichuan,China)
出处
《计算机工程》
CAS
CSCD
北大核心
2023年第11期195-202,共8页
Computer Engineering
基金
国防基础科研计划(JCKY2020404C001)
国家自然科学基金(12175187)
四川省科技计划项目(2021YFG0376)。
关键词
Γ辐射
图像增强
对比度提升
偏色校正
非线性重映射
γradiation
image enhancement
contrast improvement
color cast correction
nonlinear remapping
