基于OpenMV的图像质量监控与失真检测系统  

Image quality monitoring and distortion detection system based on OpenMV

作　　者：张学刚 王涛[2,3] 尚国清 ZHANG Xuegang;WANG Tao;SHANG Guoqing(Qinghai Minzu University,Xining 810007,China;National Demonstration Center for Experimental Communication Engineering Education,Qinghai Minzu University,Xining 810007,China;Key Laboratory of Communication Engineering,Qinghai Minzu University,Xining 810007,China)

机构地区：[1]青海民族大学,青海西宁810007 [2]青海民族大学通信工程国家级实验教学示范中心,青海西宁810007 [3]青海民族大学通信工程重点实验室,青海西宁810007

出　　处：《现代电子技术》2024年第21期51-58,共8页Modern Electronics Technique

基　　金：2022年度青海省重点研发与转化计划科技援青合作专项项目(2022-QY-205)。

摘　　要：针对图像质量监控过程中出现的失真等问题,提出一种基于OpenMV的图像质量监控与失真检测系统。该系统以STM32单片机作为主控板,搭载OpenMV作为图像采集设备,通过优化OpenMV IDE软件的函数库检测方法,对图像数据进行失真检测,在保证精度达标的情况下提高了检测速度。该系统能够识别视频监控画面中雪花噪声、亮度异常、图像偏色、图像模糊和黑屏等失真类型,并在检测出失真后进行示警。实验结果表明,该系统图像失真类型检测示警平均成功率达到了85%以上,展现出了良好的准确性和实用性。能够适应不同环境和场景的监控需要。An image quality monitoring and distortion detection system based on open machine vision toolkit(OpenMV)is proposed to get rid of the distortion in the process of image quality monitoring.The SCM STM32 is used as the main control board.The OpenMV is incorporated for image capture.An optimized function library detection method within the IDE software on OpenMV is employed to identify image distortions,which enhances detection speed while maintaining accuracy standards.The system exhibits proficiency in recognizing various distortion types,including snow noise,luminance anomalies,color deviations,blurring,and black screen within video surveillance feeds,triggering an alarm upon detection.The experimental results reveal that the proposed system has a remarkable average success rate of 85%in identifying image distortions,underscoring its high accuracy and practical utility,so it can adapt to the monitoring needs of different environments and scenes.

关 键 词：质量监控 OpenMV 雪花噪声 亮度异常 图像偏色 图像模糊 黑屏 

分 类 号：TN948.6-34[电子电信—信号与信息处理]