{"title":"Contrast Enhancement Using the Laplacian-of-a-Gaussian Filter","authors":"Neycenssac F.","doi":"10.1006/cgip.1993.1034","DOIUrl":null,"url":null,"abstract":"<div><p>A method of contrast enhancement for digital images is provided. As the method uses the LoG (Laplacian-of-a-Gaussian) filter, it is not significantly sensitive to noise. It is easy to develop this linear method when the Marr-Hildreth edge-detection technique has already been applied. LoG is roughly a bandpass filter.Therefore, only a single filter is required to enhance frequencies in one range. To enhance several frequency ranges, several standard deviation coefficients σ must be chosen for LoG convolutions. Progressive contrast ameliorations are obtained in this case. A description, an implementation, an estimate of efficiency, and a comparison with other methods are presented.</p></div>","PeriodicalId":100349,"journal":{"name":"CVGIP: Graphical Models and Image Processing","volume":"55 6","pages":"Pages 447-463"},"PeriodicalIF":0.0000,"publicationDate":"1993-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1006/cgip.1993.1034","citationCount":"86","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"CVGIP: Graphical Models and Image Processing","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S1049965283710345","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 86
Abstract
A method of contrast enhancement for digital images is provided. As the method uses the LoG (Laplacian-of-a-Gaussian) filter, it is not significantly sensitive to noise. It is easy to develop this linear method when the Marr-Hildreth edge-detection technique has already been applied. LoG is roughly a bandpass filter.Therefore, only a single filter is required to enhance frequencies in one range. To enhance several frequency ranges, several standard deviation coefficients σ must be chosen for LoG convolutions. Progressive contrast ameliorations are obtained in this case. A description, an implementation, an estimate of efficiency, and a comparison with other methods are presented.