- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
Blind Noisy Image Quality Assessment Using Sub-Band Kurtosis
摘要： Noise that afflicts natural images, regardless of the source, generally disturbs the perception of image quality by introducing a high-frequency random element that, when severe, can mask image content. Except at very low levels, where it may play a purpose, it is annoying. There exist significant statistical differences between distortion-free natural images and noisy images that become evident upon comparing the empirical probability distribution histograms of their discrete wavelet transform (DWT) coefficients. The DWT coefficients of low- or no-noise natural images have leptokurtic, peaky distributions with heavy tails; while noisy images tend to be platykurtic with less peaky distributions and shallower tails. The sample kurtosis is a natural measure of the peakedness and tail weight of the distributions of random variables. Here, we study the efficacy of the sample kurtosis of image wavelet coefficients as a feature driving an extreme learning machine which learns to map kurtosis values into perceptual quality scores. The model is trained and tested on five types of noisy images, including additive white Gaussian noise, additive Gaussian color noise, impulse noise, masked noise, and high-frequency noise from the LIVE, CSIQ, TID2008, and TID2013 image quality databases. The experimental results show that the trained model has better quality evaluation performance on noisy images than existing blind noise assessment models, while also outperforming general-purpose blind and full-reference image quality assessment methods.
关键词： sub-band,discrete wavelet transform (DWT),extreme learning machine (ELM),kurtosis,Blind noisy image quality assessment
更新于2025-09-23 15:23:52
-
[IEEE 2018 2nd International Conference on Engineering Innovation (ICEI) - Bangkok (2018.7.5-2018.7.6)] 2018 2nd International Conference on Engineering Innovation (ICEI) - Diabetic retinopathy fundus image classification using discrete wavelet transform
摘要： Diabetes is an incurable disease which erodes away body slowly, this disease in becoming common and becoming a cause of social distress. The only solution to this problem is early detection of disease and take precautionary measure to keep its effects to minimum. Since it affects various parts of body, the affected organ also includes eye which is very sensitive to any kind of distress. Diabetic Retinopathy effects of diabetes on eye retina, which includes rupturing of retina blood vessels and abnormal growth of blood vessels in retina, which ultimately causes blindness. Diabetic Retinopathy can be identified by examining the retinoscopy images. In this paper, retinoscopy images were processed using wavelet transform. Wavelet coefficients extracted from the images were obtained to identify Diabetic Retinopathy. KNN and SVM were used to classify the retinoscopy images. This papers have shown remarkable improvement as compared to previous studies, with KNN at 98.16 % accuracy and SVM at 97.85 % accuracy.
关键词： sensitivity,specificity,Discrete Wavelet Transform (DWT),accuracy,KNN,Diabetic Retinopathy (DR),histogram equalization,SVM
更新于2025-09-10 09:29:36