基于自然統(tǒng)計(jì)特征分布的無(wú)參考圖像質(zhì)量評(píng)價(jià)
doi: 10.11999/JEIT151058
基金項(xiàng)目:
國(guó)家自然科學(xué)基金(60975008)，重慶市教委科學(xué)技術(shù)研究項(xiàng)目(KJ1400434)
A No-reference Image Quality Assessment Based on Distribution Characteristics of Natural Statistics
Funds:
The National Natural Science Foundation of China (60975008), Science and Technology Research Project of Chongqing Education Committee (KJ1400434)
-
摘要: 針對(duì)目前的無(wú)參考評(píng)價(jià)方法無(wú)法準(zhǔn)確反映人類(lèi)對(duì)圖像質(zhì)量的視覺(jué)感知效果，該文提出一種基于自然統(tǒng)計(jì)特征分布(DIstribution Characteristics of Natural, DICN)的無(wú)參考圖像質(zhì)量評(píng)價(jià)方法。其原理是用小波變換將圖像分解為低頻子帶和高頻子帶部分，再將高頻子帶部分分成 的小塊，提取每一子塊的幅值和信息熵，并分別計(jì)算其分布直方圖均值和斜度作為特征，利用支持向量回歸思想對(duì)特征進(jìn)行訓(xùn)練，建立5種不同失真類(lèi)型的質(zhì)量預(yù)測(cè)模型。在此基礎(chǔ)上，采用支持向量機(jī)針對(duì)圖像特征構(gòu)造分類(lèi)器并進(jìn)行失真判斷以確定不同失真的權(quán)重，結(jié)合5種失真評(píng)價(jià)模型可得到自然統(tǒng)計(jì)特征分布的無(wú)參考評(píng)價(jià)模型。實(shí)驗(yàn)結(jié)果分析表明，該算法的評(píng)價(jià)效果優(yōu)于現(xiàn)有的經(jīng)典算法，與主觀評(píng)價(jià)具有較好一致性，能夠準(zhǔn)確反映人類(lèi)對(duì)圖像質(zhì)量的視覺(jué)感知效果。
-
關(guān)鍵詞:
- 無(wú)參考圖像質(zhì)量評(píng)價(jià) /
- 小波分解 /
- 局部幅值 /
- 局部熵
Abstract: The current No-Reference Image Quality Assessment (NR-IQA) methods are not well consistent with subjective evaluation, a novel NR-IQA method based on the DIstribution Characteristics of Natural statistics (DICN) is proposed in this paper. In the proposed method, image is decomposed into low frequency subbands and high frequency subbands with wavelet, and its high frequency subbands are divided into blocks at size of 88, their amplitude and entropy are respectively extracted from the blocks, then their mean values of the distribution histogram and skewness are respectively calculated, and their results are as the image features. The features trained by Support Vector Regression (SVR) are for building 5 kinds of distortion image quality pre-measurement model. To determine the weights of the different distortions, the image features of classifier based on SVR are structured for carrying out the distortion evalution. Based on 5 kinds of distortion evaluation models, the NR-IQA model with the natural statistical distribution can be obtained. The results of experiments show that the proposed method performance is better than the present classical methods. The method is well consistent with the subjective assessment results, and can reflect human subjective feeling well. -
REDZUAN A M and LING Shao. Non-distortion-specific no-reference image quality assessment: a survey[J]. Information Sciences, 2015, 301(4): 141-160. doi: 10.1016/ j.ins.2014.12.055. 賈惠珍, 孫權(quán)森, 王同罕. 結(jié)合感知特征和自然場(chǎng)景統(tǒng)計(jì)的無(wú)參考圖像質(zhì)量評(píng)價(jià)[J]. 中國(guó)圖象圖形學(xué)報(bào), 2014, 19(6): 859-867. doi: 10.11834/jig.20140606. JIA Huizhen, SUN Quansen, and WANG Tonghan. Blind image quality assessment based on perceptual features and natural scene statistics[J]. Journal of Image and Graphics, 2014, 19(6): 859-867. doi: 10.11834/jig.20140606. MOORTHY A K and BOVIK A C. A two-step framework for constructing blind image quality indices[J]. IEEE Signal Processing Letters, 2010, 17(5): 513-516. doi: 10.1109/LSP. 2010.2043888. MOORTHY A K and BOVIK A C. Blind image quality assessment: From natural scene statistics to perceptual quality[J]. IEEE Transactions on Image Processing, 2011, 20(12): 3350-3364. doi: 10.1109/TIP.2011.2147325. HE L, TAO D, LI X, et al. Sparse representation for blind image quality assessment[C]. 25th IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Rhode Island, 2012: 1146-1153. doi: 10.1109/CVPR.2012.6247795. 金波, 李朝鋒, 吳小俊. 結(jié)合NSS和小波變換的無(wú)參考圖像質(zhì)量評(píng)價(jià)[J]. 中國(guó)圖象圖形學(xué)報(bào), 2012, 17(1): 33-39. JIN Bo, LI Chaofeng, and WU Xiaojun. No-reference image quality assessment based on natural scene statistics and wavelet[J]. Journal of Image and Graphics, 2012, 17(1): 33-39. SAAD M A, BOVIK A C, and CHARRIER C. DCT statistics model-based blind image quality assessment[C]. IEEE International Conference on Image Processing, Brussels, Belgium, 2011: 3093-3096. doi: 10.1109/ICIP.2011.6116319. MITTAL A, MOORTHY A K, and BOVIK A C. No-reference image quality assessment in the spatial domain [J]. IEEE Transactions on Image Processing, 2012, 21(12): 4695-4708. doi: 10.1109/TIP.2012.2214050. MITTAL A, SOUNDARARAN R, and BOVIK A C. Making a completely blind image quality analyzer[J]. IEEE Signal Processing Letters, 2013, 20(3): 209-212. doi: 10.1109/ LSP.2012.2227726. 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, and 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. VAPNIK V N. An overview of statistical learning theory [J]. IEEE Transactions on Neural Networks, 1999, 10(5): 988-999. doi: 10.1109/72.788640. 付曉薇, 代蕓, 陳黎, 等. 基于局部熵的量子衍生醫(yī)學(xué)超聲圖像去斑[J]. 電子與信息學(xué)報(bào), 2015, 37(3): 560-566. doi: 10. 11999/JEIT140587. FU Xiaowei, DAI Yun, CHEN Li, et al. Quantum-inspired despeckling of medical ultrasound images based on local entropy[J]. Journal of Electronics Information Technology, 2015, 37(3): 560-566. doi: 10.11999/JEIT140587. LIU Lixiong, LIU Bao, HUANG Hua, et al. No-reference image quality assessment based on spatial and spectral entropies[J]. IEEE Transactions on Signal Processing: Image Communication, 2014, 29(8): 856-863. doi: 10.1016/j.image. 2014.06.006. SCHLKOPF B, SMOLA A J, WILLIAMSON R C, et al. New support vector algorithms[J]. Neural Computation, 2000, 12(5): 1207-1245. doi: 10.1162/089976600300015565. 周曉劍. 考慮梯度信息的支持向量回歸機(jī)[J]. 自動(dòng)化學(xué)報(bào), 2014, 40(12): 2908-2915. doi: 10.3724/SP.J.1004.2014.02908. ZHOU Xiaojian. Enhancingsupport vector regression with gradient information[J]. Acta Automatica Sinica, 2014, 40(12): 2908-2915. doi: 10.3724/SP.J.1004.2014.02908. 張繼紅, 鄭俊生. 多元再生核徑向基函數(shù)研究[J]. 大連交通大學(xué)學(xué)報(bào), 2015, 36(1): 109-117. doi: 10.13291/j.cnki.djdxac. 2015.01.029. ZHANG Jihong and ZHENG Junsheng. Study of multivariate reproducing kernel radial basis function[J]. Journal of Dalian Jiaotong University, 2015, 36(1): 109-117. doi: 10.13291/ j.cnki.djdxac.2015.01.029. SHEIKH H R and BOVIK A C. Image information and visual quality[J]. IEEE Transactions on Image Processing, 2006, 15(2): 430-444. doi: 10.1109/TIP.2005.859378. SHEIKH H R, BOVIK A C, and DE VECIANA G. An information fidelity criterion for image quality assessment using natural scene statistics[J]. IEEE Transactions on Image Processing, 2005, 14(12): 2117-2128. doi: 10.1109/TIP.2005. 859389. -
計(jì)量
- 文章訪問(wèn)數(shù): 1331
- HTML全文瀏覽量: 111
- PDF下載量: 627
- 被引次數(shù): 0
下載:
