无参考样本下的食管内镜图像增强技术  

Esophageal endoscopic image enhancement method without reference samples

作　　者：姚韩敏 周颖玥[1,2] 郭俊菲 秦佳敏 李小霞[1,2] 董舒琦[1] Yao Hanmin;Zhou Yingyue;Guo Junfei;Qin Jiamin;Li Xiaoxia;Dong Shuqi(School of Information Engineering,Southwest University of Science and Technology,Mianyang 621010,China;Robot Technology Used for Special Environment Key Laboratory of Sichuan Province,Mianyang 621010,China;Sichuan Mianyang 404 Hospital,Mianyang 621053,China)

机构地区：[1]西南科技大学信息工程学院,绵阳621010 [2]特殊环境机器人技术四川省重点实验室,绵阳621010 [3]四川绵阳四〇四医院,绵阳621053

出　　处：《中国图象图形学报》2024年第11期3487-3500,共14页Journal of Image and Graphics

基　　金：四川省科技计划资助(2023YFG0262);2023年四川省大学生创新创业训练计划项目(S202310619102)。

摘　　要：目的在食管病变的筛查中,卢戈染色内镜(Lugol’s chromo endoscopy,LCE)因其良好的病变可视性、诊断准确性以及低廉的检查成本在消化内科检查中独具优势。然而,在采集LCE食管内镜图像时,由于内窥镜内置光源的限制,光照的方向和角度有限,导致图像出现光照不均匀、对比度低等问题。方法针对这一问题,本文在Ret⁃inexDIP算法基础上,提出了用于生成图像分量的生成器网络(stable generating network,SgNet)。该网络采用编码—解码结构,通过本文提出的通道调整模块(channel attention adjustment module,CAAM)使得上下采样过程中对应的特征通道权重保持一致,以增强网络稳定性,进而提升生成图像的质量。同时提出了一种新的颜色模型——“固定比例、独立亮度”模型(fixed proportion light,FPL),该模型将图像的亮度信息和颜色比例信息独立表示出来,图像的光照增强过程只在亮度通道上进行调整,从而保证LCE食管内镜图像的整体色彩信息不紊乱。结果在自建的LCE低光图像数据集上测试本文算法的有效性,与多种主流低光图像增强算法进行视觉效果和客观指标评价比较。结果显示本文所提算法在颜色保真、对比度提升以及降低噪声干扰等方面具有优势,在自然图像质量评估器(natu⁃ral image quality evaluator,NIQE)和盲/无参考空间图像质量评估器(blind/referenceless image spatial quality evaluator,BRISQUE)指标上均表现出色。结论综合来看,本文算法在增强LCE食管内镜图像亮度的同时,有效地保持了图像的色彩和纹理细节信息,可以帮助医生更清晰地观察病灶组织结构和细节,提升诊断准确率,并为后续病灶智能检测提供了优质的图像数据。Objective Esophageal cancer is one of the most common malignant tumors that seriously threaten human health.At present,endoscopy combined with histopathological biopsy is the“gold standard”for diagnosing early esophageal cancer.Among them,Lugol’s chromo endoscopy(LCE)has a unique advantage in gastroenterology because of its good lesion visibility,diagnostic accuracy,and low cost.However,with the rising number of patients,the imbalance between the number of doctors and patients is becoming increasingly serious.The manual diagnosis process based on endoscopic images is susceptible to several factors,such as the experience and mental state of the doctor,the limitation of diagnosis time,the enormous image base,and the complex and variable appearance of the lesion.Therefore,the clinical diagnosis of artificial esophageal lesions still has a high rate of missed diagnoses and misdiagnosis.In recent years,the application of artificial intelligence(AI)in the field of medical imaging has provided strong support for doctors,and the AIassisted diagnosis system based on deep learning can assist doctors to accurately diagnose the location and type of lesions,reducing their burden.However,deep learning models need sufficient and high-quality data.For esophageal endoscopic images,LCE esophageal endoscopic images will inevitably be affected by the built-in light source of the acquisition device during the acquisition process.The light distribution of LCE esophageal endoscopic images will be uneven due to the limited illumination direction and angle of the built-in light source,affecting the overall quality of the images,which is unfavorable to the subsequent training of the intelligent lesion detection model.The existing low-light image enhancement algorithms are not ideal for the enhancement of LCE esophageal endoscopic images due to the special nature of LCE esophageal endoscopic images,complex illumination,color sensitivity,and lack of high-quality reference(paired or unpaired)datasets.Method Based on the“generative”

关 键 词：图像增强 卢戈染色内镜(LCE) Retinex模型 图像生成 颜色模型 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]