https://scholars.lib.ntu.edu.tw/handle/123456789/607204
標題: | Pixelwise Image Sharpness Based on the Weighted Response Ratios of Short and Long Edge Detectors | 作者: | Tsai Y.-W. JIAN-JIUN DING |
關鍵字: | blurring;edge detector;image metric;sharpness;Image quality;Blurring;Depth Estimation;Edge detectors;Foreground extraction;Image metrics;Image quality measurement;Image sharpness;Longest edge;Measurement methods;Sharpness;Image enhancement | 公開日期: | 2021 | 起(迄)頁: | 107-108 | 來源出版物: | 2021 IEEE 10th Global Conference on Consumer Electronics, GCCE 2021 | 摘要: | Image sharpness is a metric to determine whether a region within an image is clear or blurring. It is helpful for image quality measurement, foreground extraction, image enhancement, depth estimation, and adaptive image processing. However, for most of existing sharpness measurement methods, it costs much computation time to compute the sharpness locally and pixelwisely. In this paper, we propose an effective and efficient image sharpness measurement method. Its idea is based on that, for a blurred edge, the response of the short-length edge detector is much smaller than that of the long-length edge detector. By contrast, for a sharp edge, the difference is relative smaller. Experiments show that, with the proposed method, the blurred and the clear regions can be well distinguished. Moreover, it requires much less computation loading and can be performed in real time. ? 2021 IEEE. |
URI: | https://www.scopus.com/inward/record.uri?eid=2-s2.0-85123490349&doi=10.1109%2fGCCE53005.2021.9622011&partnerID=40&md5=c44ec0bc61b745eafb1685ead10f3eec https://scholars.lib.ntu.edu.tw/handle/123456789/607204 |
DOI: | 10.1109/GCCE53005.2021.9622011 |
顯示於: | 電機工程學系 |
在 IR 系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。