面向圖像集的置信區(qū)間內(nèi)一致性增強(qiáng)質(zhì)量評(píng)價(jià)準(zhǔn)則

劉浩; 孫曉帆; 張鑫生; 吳樂明; 況奇剛

doi:10.11999/JEIT180272

面向圖像集的置信區(qū)間內(nèi)一致性增強(qiáng)質(zhì)量評(píng)價(jià)準(zhǔn)則

doi: 10.11999/JEIT180272 cstr: 32379.14.JEIT180272

東華大學(xué)信息科學(xué)與技術(shù)學(xué)院 ??上海 ??201620

基金項(xiàng)目: 上海市自然科學(xué)基金(18ZR1400300)

詳細(xì)信息

作者簡介:
劉浩：男，1977年生，副教授，研究方向?yàn)閳D像處理與通信

孫曉帆：女，1993年生，碩士生，研究方向?yàn)樗聢D像增強(qiáng)及質(zhì)量評(píng)價(jià)

張鑫生：男，1992年生，碩士生，研究方向?yàn)閴嚎s域圖像分析

吳樂明：男，1993年生，碩士生，研究方向?yàn)閳D像集壓縮及評(píng)價(jià)

況奇剛：男，1994年生，碩士生，研究方向?yàn)閳D像超分辨率

通訊作者:
劉浩　liuhao@dhu.edu.cn

中圖分類號(hào): TP391
計(jì)量
- 文章訪問數(shù): 1982
- HTML全文瀏覽量: 578
- PDF下載量: 76
- 被引次數(shù): 0
出版歷程
- 收稿日期: 2018-03-23
- 修回日期: 2018-09-05
- 網(wǎng)絡(luò)出版日期: 2018-09-19
- 刊出日期: 2019-01-01

Consistency Enhancement Quality Assessment Criterion in Confidence Interval for Image Set

College of Information Science and Technology, Donghua University, Shanghai 201620, China

Funds: The Natural Science Foundation of Shanghai (18ZR1400300)

摘要

摘要:
在對(duì)整個(gè)圖像集進(jìn)行增強(qiáng)質(zhì)量評(píng)價(jià)時(shí)，現(xiàn)有的平均準(zhǔn)則會(huì)隨著不同圖像集非一致性地變化，從而導(dǎo)致較大的評(píng)價(jià)質(zhì)量波動(dòng)。為此，該文提出一個(gè)面向圖像集的置信區(qū)間內(nèi)一致性增強(qiáng)質(zhì)量評(píng)價(jià)準(zhǔn)則，通過設(shè)置應(yīng)用參數(shù)并使用置信區(qū)間篩選數(shù)據(jù)，再比較各圖像增強(qiáng)前后的質(zhì)量分?jǐn)?shù)差值，由此評(píng)估圖像質(zhì)量增強(qiáng)的一致性，最終計(jì)算出一致性增強(qiáng)質(zhì)量分?jǐn)?shù)有效值。在眾多圖像增強(qiáng)算法中，所提準(zhǔn)則能夠挑選出具體應(yīng)用所需要的穩(wěn)定性強(qiáng)、可靠性高的增強(qiáng)算法。實(shí)驗(yàn)結(jié)果表明，所提準(zhǔn)則具有良好的主客觀評(píng)價(jià)一致性，性能優(yōu)于當(dāng)前的平均準(zhǔn)則，為各種圖像增強(qiáng)算法提供了一個(gè)可用于任意圖像集的質(zhì)量評(píng)價(jià)準(zhǔn)則。
- 圖像增強(qiáng) /
- 圖像集 /
- 質(zhì)量評(píng)價(jià) /
- 一致性增強(qiáng) /
- 置信區(qū)間
Abstract:
When evaluating the enhancement quality of a whole image set, the existing average score criterion varies inconsistently with different image sets and produces a large evaluation quality fluctuation. Therefore, this paper proposes a consistency enhancement quality assessment criterion in confidence interval for any image set. By setting application parameters and using confidence interval to screen data, the proposed criterion compares the quality score difference before and after enhancing each image, and evaluates the consistency of image quality enhancement, and then calculates the effective value of consistency enhancement quality scores. Among many image enhancement algorithms, the proposed criterion can select the high-reliability enhancement algorithm for a specific application. The experimental results show that the proposed criterion has good subjective and objective consistency and outperforms the existing average score criterion, which provides an evaluation criterion for those image enhancement algorithms applied to any image set.
- Image enhancement /
- Image set /
- Quality assessment /
- Consistency enhancement /
- Confidence interval

HTML全文

圖 1 不同圖像增強(qiáng)算法的定性比較

下載: 全尺寸圖片幻燈片

圖 2 各增強(qiáng)算法下的質(zhì)量分?jǐn)?shù)曲線

下載: 全尺寸圖片幻燈片

圖 3 典型水下圖像

下載: 全尺寸圖片幻燈片

圖 4 圖像增強(qiáng)前后的熵值質(zhì)量分?jǐn)?shù)對(duì)比

下載: 全尺寸圖片幻燈片

圖 5 CEQA準(zhǔn)則的質(zhì)量評(píng)價(jià)體系

下載: 全尺寸圖片幻燈片

圖 6 CEQA準(zhǔn)則的流程圖

下載: 全尺寸圖片幻燈片

圖 7 各幅圖像的圖像質(zhì)量分?jǐn)?shù)

下載: 全尺寸圖片幻燈片

表 1 熵值的平均分?jǐn)?shù)對(duì)比

算法	原圖	逆濾波^[6]	均值濾波^[7]	動(dòng)態(tài)閾值白平衡^[8]	直方圖均衡^[9]	CLAHE-RGB^[10]	對(duì)比度拉伸^[11]	無監(jiān)督顏色模型^[12]
熵值	7.41	7.47	7.38	7.18	7.49	7.72	7.21	7.55

下載: 導(dǎo)出CSV

表 2 平均準(zhǔn)則的平均值與CEQA準(zhǔn)則的${\rm{CEQ}}{{\rm{A}}_{{\rm{eff}}}}$值的對(duì)比(在UCIQE方法下)

	圖像集	原圖	動(dòng)態(tài)閾值白平衡^[8]	對(duì)比度拉伸^[11]
平均值	圖像集a	0.422	0.427	0.426
	圖像集b	0.418	0.408	0.433
	圖像集c	0.419	0.389	0.446
${\rm{CEQ}}{{\rm{A}}_{{\rm{eff}}}}$值	圖像集a	無	–0.101	0.698
	圖像集b		–0.510	0.294
	圖像集c		–0.792	0.199

下載: 導(dǎo)出CSV

表 3 熵值的${\rm{CEQ}}{{\rm{A}}_{{\rm{eff}}}}$值

增強(qiáng)算法	CLAHE-RGB^[10]	逆濾波^[6]	無監(jiān)督顏色模型^[12]	直方圖均衡^[9]	動(dòng)態(tài)閾值白平衡^[8]	對(duì)比度拉伸^[11]	均值濾波^[7]
${\rm{CEQ}}{{\rm{A}}_{{\rm{eff}}}}$值	0.202	0.193	–0.025	–0.205	–0.609	–0.713	–0.8057

下載: 導(dǎo)出CSV

參考文獻(xiàn)(16)

王志明. 無參考圖像質(zhì)量評(píng)價(jià)綜述[J]. 自動(dòng)化學(xué)報(bào), 2015, 41(6): 1062–1079. doi: 10.16383/j.aas.2015.c140404

WANG Zhiming. Review of no-reference image quality assessment[J]. Acta Automatica Sinica, 2015, 41(6): 1062–1079. doi: 10.16383/j.aas.2015.c140404

PENG Y T and COSMAN P C. Underwater image restoration based on image blurriness and light absorption[J]. IEEE Transactions on Image Processing, 2017, 26(4): 1579–1594. doi: 10.1109/TIP.2017.2663846

LI Chongyi, GUO Jichang, CONG Runmin, et al. Underwater image enhancement by dehazing with minimum information loss and histogram distribution prior[J]. IEEE Transactions on Image Processing, 2016, 25(12): 5664–5677. doi: 10.1109/TIP.2016.2612882

LI Chongyi, GUO Jichang, CHEN Shanji, et al. Underwater image restoration based on minimum information loss principle and optical properties of underwater imaging[C]. IEEE International Conference on Image Processing, Phoenix, USA, 2016: 1993–1997.

YANG M and SOWMYA A. An underwater color image quality evaluation metric[J]. IEEE Transactions on Image Processing, 2015, 24(12): 6062–6071. doi: 10.1109/TIP.2015.2491020

吳雪垠, 吳謹(jǐn), 張鶴. 逆濾波法在圖像復(fù)原中的應(yīng)用[J]. 信息技術(shù), 2011(10): 183–185. doi: 10.3969/j.issn.1009-2552.2011.10.050

WU Xueyin, WU Jin, and ZHANG He. Research on image restoration techniques based on inverse filtering algorithm[J]. Information Technology, 2011(10): 183–185. doi: 10.3969/j.issn.1009-2552.2011.10.050

何石, 潘曉璐, 李一民. 一種均值濾波的優(yōu)化算法[J]. 信息技術(shù), 2012(3): 133–137. doi: 10.3969/j.issn.1009-2552.2012.03.038

HE Shi, PAN Xiaolu, and LI Yimin. Optimization algorithm for average filtering[J]. Information Technology, 2012(3): 133–137. doi: 10.3969/j.issn.1009-2552.2012.03.038

蘇志鋒. 基于FPGA的圖像預(yù)處理研究與實(shí)現(xiàn)[D]. [博士論文], 華南理工大學(xué), 2015.

SU Zhifeng. Studying and implementation of image signal preprocessing based on FPGA[D]. [Ph.D. dissertation], South China University of Technology, 2015.

李耀輝, 劉保軍. 基于直方圖均衡的圖像增強(qiáng)[J]. 華北科技學(xué)院學(xué)報(bào), 2003, 5(2): 65–67.

LI Yaohui and LIU Baojun. The image enhancement based on histogram equalization[J]. Journal of North China Institute of Science and Technology, 2003, 5(2): 65–67.

HITAM M S, AWALLUDIN E A, and YUSSOF W. Mixture contrast limited adaptive histogram equalization for underwater image enhancement[C]. International Conference on Computer Application Technology, Sousse, Tunisia, 2013: 1–5.

陳宇, 霍富榮, 苗華. 對(duì)比度拉伸在目標(biāo)探測與識(shí)別中的應(yīng)用研究[J]. 儀器儀表學(xué)報(bào), 2008, 29(4): 795–798.

CHEN Yu, HUO Furong, and MIAO Hua. Application of contrast stretching in optical correlation detection and recognition[J]. Chinese Journal of Scientific Instrument, 2008, 29(4): 795–798.

楊勇, 郭玲, 王天江. 基于多尺度結(jié)構(gòu)張量的多類無監(jiān)督彩色紋理圖像分割方法[J]. 計(jì)算機(jī)輔助設(shè)計(jì)與圖形學(xué)學(xué)報(bào), 2014, 26(5): 812–825.

YANG Yong, GUO Ling, and WANG Tianjiang. Multi-scale structure tensor based unsupervised color-texture image segmentation approach in multiclass[J]. Journal of Computer-Aided Design &Computer Graphics, 2014, 26(5): 812–825.

蔣剛毅, 黃大江, 王旭, 等. 圖像質(zhì)量評(píng)價(jià)方法研究進(jìn)展[J]. 電子與信息學(xué)報(bào), 2010, 32(1): 219–226. doi: 10.3724/SP.J.1146.2009.00091

JIANG Gangyi, HUANG Dajiang, WANG Xu, et al. Overview on image quality assessment methods[J]. Journal of Electronics &Information Technology, 2010, 32(1): 219–226. doi: 10.3724/SP.J.1146.2009.00091

EMBERTON S, CHITTKA L, and CAVALLARO A. Hierarchical rank-based veiling light estimation for underwater dehazing[C]. British Machine Vision Conference, Swansea, UK, 2015: 125.1–125.12.

JIAN Muwei, QI Qiang, DONG Junyu, et al. The OUC-Vision large-scale underwater image database[C]. IEEE International Conference on Multimedia & Expo, Hong Kong, China, 2017: 1297–1302.

JIAN Muwei, QI Qiang, DONG Junyu, et al. Saliency detection using quaternionic distance based weber local descriptor and level priors[J]. Multimedia Tools and Applications, 2018, 77(11): 14343–14360. doi: 10.1007/s11042-017-5032-z

相關(guān)文章

施引文獻(xiàn)

資源附件(0)

訪問統(tǒng)計(jì)