An Automated Tool to Remove Flash Light Spots using Image Processing Techniques

Purohit Shrinivasacharya, Jagadamba G, Suresh Jagadeeshaiah
{"title":"An Automated Tool to Remove Flash Light Spots using Image Processing Techniques","authors":"Purohit Shrinivasacharya, Jagadamba G, Suresh Jagadeeshaiah","doi":"10.1109/ICICICT54557.2022.9917639","DOIUrl":null,"url":null,"abstract":"An image is captured using a flash list on reflecting surfaces or glass objects, and it causes a bright light on a particular spot in an image. It is because of the direct reflection of flash resulting from glassy objects and results in loss of data over the reflected part of an image. To overcome these problems, an application is developed to capture and process the images. This application can capture the images in a single click in both with flash and without flash condition. The captured images are divided into a small number of blocks and calculate Peak Signal to Noise Ratio (PSNR) value for each block. Based on these PSNR values the blocks of a flash image is replaced with blocks without flash image and while replacement image intensity of an image will change to the same as the flash image. The resultant image is free from a flashlight spot and lost information. This technique is tested on different images and tabulated the corresponding values for comparing the result. The proposed method provides intuitively better results than the pixel-wise method. This application will be used in the photography to remove the reflected objects' flashlight spot.","PeriodicalId":246214,"journal":{"name":"2022 Third International Conference on Intelligent Computing Instrumentation and Control Technologies (ICICICT)","volume":"62 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-08-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 Third International Conference on Intelligent Computing Instrumentation and Control Technologies (ICICICT)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICICICT54557.2022.9917639","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

An image is captured using a flash list on reflecting surfaces or glass objects, and it causes a bright light on a particular spot in an image. It is because of the direct reflection of flash resulting from glassy objects and results in loss of data over the reflected part of an image. To overcome these problems, an application is developed to capture and process the images. This application can capture the images in a single click in both with flash and without flash condition. The captured images are divided into a small number of blocks and calculate Peak Signal to Noise Ratio (PSNR) value for each block. Based on these PSNR values the blocks of a flash image is replaced with blocks without flash image and while replacement image intensity of an image will change to the same as the flash image. The resultant image is free from a flashlight spot and lost information. This technique is tested on different images and tabulated the corresponding values for comparing the result. The proposed method provides intuitively better results than the pixel-wise method. This application will be used in the photography to remove the reflected objects' flashlight spot.
使用图像处理技术自动删除闪光点的工具
在反射表面或玻璃物体上使用闪光表捕捉图像,并在图像中的特定点上产生明亮的光。这是因为由玻璃状物体引起的闪光的直接反射,导致图像反射部分的数据丢失。为了克服这些问题,开发了一个应用程序来捕获和处理图像。这个应用程序可以捕捉图像在一个单一的点击在闪光灯和没有闪光灯条件。将捕获的图像分成小块,计算每个块的峰值信噪比(PSNR)值。基于这些PSNR值,将flash图像的块替换为没有flash图像的块,同时图像的替换图像强度将改变为与flash图像相同。生成的图像不受手电筒斑点和丢失信息的影响。该技术在不同的图像上进行了测试,并将相应的值制成表格以比较结果。该方法在直观上优于逐像素方法。这个应用程序将在摄影中用于去除反射物体的手电筒斑点。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信