4.3. Saliency-based multi-feature modeling for semantic image retrieval / Cong Bai, Jia-nan Chen, Ling Huang, Kidiyo Kpalma, Shengyong Chen // Journal of Visual Communication and Image Representation, Volume 50, January 2018, Pages 199-204. – https://ac. /S1047320317302304/1-s2.0-S1047320317302304-main. pdf?_tid=abaf6ee4-ebc7-11e7-a0b5-00000aab0f02&acdnat=1514463027_9526a0925fbe5a9b368696bafe19e44c
Semantic gap is an important challenging problem in content-based image retrieval (CBIR) up to now. Bag-of-words (BOW) framework is a popular approach that tries to reduce the semantic gap in CBIR. In this paper, an approach integrating visual saliency model with BOW is proposed for semantic image retrieval. Images are firstly segmented into background regions and foreground objects by a visual saliency-based segmentation method. And then multi-features including Scale Invariant Feature Transform (SIFT) features packed in BOW are extracted from regions and objects respectively and fused considering different characteristics of background regions and foreground objects. Finally, a fusion of z-score normalized Chi-Square distance is adopted as the similarity measurement.
4.4. Improved field of experts model for image completion / Yueli Li, Hao Wu, Liangchi Li, Yuqi Yang, Rongfang Bie // Optik - International Journal for Light and Electron Optics, Volume 142, August 2017, Pages 174-182. - https://ac. /S1077314215000235/1-s2.0-S1077314215000235-main. pdf?_tid=8bc9ba3c-ebc5-11e7-969b-00000aab0f6b&acdnat=1514462115_cd3ae76a4b22ccdf92556b4d6d764126
The way to complete such missing regions effectively has thus become an important topic in recent years and has attracted the attention of many researchers. However, existing methods work satisfactorily only when the missing region is consistent and regular. If the missing pixels are discrete, then the completion results are unrealistic. Moreover, the amount of calculation becomes burdensome. To solve the aforementioned problem, we propose an improved field of experts model for image completion. On the one hand, we use a wavelet transform model to optimize the completion results. On the other hand, we use a best learning image selection model to improve the completion process.
Generative model based image annotation methods have achieved good annotation performance. However, due to the problem of “semantic gap”, these methods always suffer from the images with similar visual features but different semantics. It seems promising to separate these images from semantic relevant ones by using discriminant models, since they have shown excellent generalization performance. Motivated to gain the benefits of both generative and discriminative approaches, in this paper, we propose a novel image annotation approach which combine the generative and discriminative models through local discriminant topics in the neighborhood of the unlabeled image. Singular Value Decomposition(SVD) is applied to group the images of the neighborhood into different topics according to their semantic labels. The semantic relevant images and the irrelevant ones are always assigned into different exploiting the discriminant information between different topics, Support Vector Machine(SVM) is applied to classify the unlabeled image into the relevant topic, from which the more accurate annotation will be obtained by reducing the bad influence of irrelevant images.
4.6. Zhong-jie Zhu, Yu-er Wang, Gang-yi Jiang. Unsupervised segmentation of natural images based on statistical modeling // Neurocomputing, Volume 252, 23 August 2017, Pages 95-101. - https://ac. /S0925231217306549/1-s2.0-S0925231217306549-main. pdf?_tid=9d3994ac-de89-11e7-b992-00000aacb361&acdnat=1513007009_e067672b74225a693415af89f825d146
A novel unsupervised scheme for natural image segmentation is proposed aiming to acquire perceptually consistent results. Firstly, comprehensive visual features besides raw color values are extracted, including spatial frequency, contrast sensitivity, color deviation, and so on. Secondly, high correlations among visual features are reduced via principal component analysis (PCA) and the raw image pixels are then converted to a collection of feature vectors in a multi-dimensional feature space. Thirdly, the Gaussian mixture model (GMM) is employed to approximate the class distribution of image pixels and an improved expectation maximization (EM) algorithm is introduced to estimate model parameters. Finally, segmentation results are obtained by grouping of pixels based on the mixture components.
4.7. Bo Fu, Wei-Wei Li, You-Ping Fu, Chuan-Ming Song. An image topic model for image denoising // Neurocomputing, Volume 169, 2 December 2015, Pages 119-123. - https://ac. /S0925231215006761/1-s2.0-S0925231215006761-main. pdf?_tid=e782f100-de86-11e7-ab6c-00000aacb361&acdnat=1513005845_0185bc588f270c4a9f64ce9572fd7d7a.
We introduce a novel image topic model, called Latent Patch Model (LPM), which is a generative Bayesian model and assumes that the image and pixels are connected by a latent patch layer. Based on the LPM, we further propose an image denoising algorithm namely multiple estimate LPM (MELPM). Unlike other works, the proposed denoising framework is totally implemented on the latent patch layer, and it is effective for both Gaussian white noises and impulse noises.
Image formation model of scattering is proposed to enhance color retinal images in this paper. Two parameters of this model, background illuminance and transmission map, are estimated based on extracted background and foreground. The complex nature of the foreground of a retinal image, involving pixels with both low and high intensity, posed a challenge to the proper extraction of these pixels. Therefore, a new method combining Mahalanobis distance discrimination and global spatial entropy-based contrast enhancement is proposed to extract foreground pixels. It extracts background and foreground in high intensity region and low intensity region respectively and it can perform well in blurry image with tiny intensity range.
4.9. Lei Wang, Yan Chang, Hui Wang, Zhenzhou Wu, Xiaodong Yang. Information Sciences, Volumes 418–419, December 2017, Pages 61-73
Active contour models are popular and widely used for a variety of image segmentation applications with promising accuracy, but they may suffer from limited segmentation performances due to the presence of intensity inhomogeneity. To overcome this drawback, a novel region-based active contour model based on two different local fitted images is proposed by constructing a novel local hybrid image fitting energy, which is minimized in a variational level set framework to guide the evolving of contour curves toward the desired boundaries. The proposed model is evaluated and compared with several typical active contour models to segment synthetic and real images with different intensity characteristics. Experimental results demonstrate that the proposed model outperforms these models in terms of accuracy in image segmentation.
4.10. Anton David Brink. Image models and the definition of image entropy applied to the problem of unsupervised segmentation. Thesis PhD, Johannesburg, March 1994 // http://mobile. wiredspace. wits. ac. za/bitstream/handle/10539/20956/Brink%20Anton%20David._Image%20models%20and%20the%20definition. pdf? sequence=1&isAllowed=y.
This thesis attempts to clarify the notion of image entropy to which end various image and image segmentation models are discussed and proposed.
4.11. Aleksandra Piћurica and Wilfried Philips. Multiscale Statistical Image Models and Bayesian Methods // https://telin. ugent. be/~sanja/Papers/MultiscaleSPIE03.pdf.
We discuss different multiscale statistical image models in the framework of Bayesian image denoising. We discuss two important problems in specifying priors for image wavelet coefficients. The first problem is the characterization of the marginal subband statistics. Different existing models include highly kurtotic heavy-tailed distributions, Gaussian scale mixture models and weighted sums of two different distributions. We discuss the choice of a particular prior and give some new insights in this problem. The second problem that we address is statistical modelling of inter - and intrascale dependencies between image wavelet coefficients. Here we discuss the use of Hidden Markov Tree models, which are efficient in capturing inter-scale dependencies, as well as the use of Markov Random Field models, which are more efficient when it comes to spatial (intrascale) correlations. Apart from these relatively complex models, we review within a new unifying framework a class of low-complexity locally adaptive methods, which encounter the coefficient dependencies via local spatial activity indicators.
4.12. ПОСТРОЕНИЕ МОДЕЛИ ЦИФРОВОГО ИЗОБРАЖЕНИЯ ДЛЯ РЕШЕНИЯ ЗАДАЧ СЕГМЕНТАЦИИ // Системи обробки інформації, 2009, випуск 6 (80) ISSN 1681-7710. – С. 112-115. - https://yandex. ru/clck/jsredir? bu=uniq1512724627680435735&from=yandex. ru%3Bsearch%2F%3Bweb%3B%3B&text=&etext=1629.O6wFoLTIJ04rC741F0MUIu6l8RAYZYv8zJskOEb5JIbhMKP8rYmdtR301zfRCq3mMnpqIhYhhaYQRrr_jGSMnw. bafa7ecd09c8c9d84c8aea3f76f5615a36f6f14a&uuid=&state=PEtFfuTeVD5kpHnK9lio9T6U0-imFY5IshtIYWJN7W-V64A9Yd8Kv-PJgis4UdqY898U4_M9m95WwUcanXFu37Y6OOxauOn1Ir561ZEe-iQ,&&cst=AiuY0DBWFJ5fN_r-AEszk2KVb2s-qD6z8yiENAlpD3zGi5KwJ1mXdEr5EVGNj2VB7WIPnPZaQtuarJsQ4OEDyt6dP7tlZ6jY0TQ3thDEg1aYSivL9XXD0tycOPSs-88fWu2ilBjjuX6sv7ulRFji3xVTHCMXegXXgiSDg2AcPpp-v65pSTbmdR20W9NCF2pzR_wlKBmnoZRnAJ9mh7aWnLnXiUueNrRexfK2PrYF1ZL0PwXPVw7rgQ,,&data=UlNrNmk5WktYejR0eWJFYk1Ldmtxbm5oNDBPTVAzQmtYcjR5b3lfeG5pUVJxNmpTM0hyMHlxaFplLUExTk1oX1d1cFB0MFhzc1pORXBsVFdIbXk1aTV0U0t1dEttMVFCZFc0ejNxYWVNWFcxampFejZFeU5IaGMxY2FtQ2lOZ29BWlRiaGNWZEZCUE1RMXk2R0FYeFhPbTBJVDR1c3NxTVBpUllaZGpta0ZvLA,,&sign=2ca82828714b506f8be76b6a6bb618a6&keyno=0&b64e=2&ref=orjY4mGPRjk5boDnW0uvlrrd71vZw9kp5uQozpMtKCWbgDHRNa-0yra-LU08xCWQ1nAuViHDr9kxOWfQXZ8OFPH318ss07F7z2VGSdp0OaXbDaTo6BKFXTRGPw6CV4RpCr9CochuwvpN4ZQJLeqIq23IGqefRnQziR62CNDFEu6GG-ZpuZ2kc2Fpp4ePd7tHEaqLyfndj_bv5gj0hdgsig8Sqc_odA24Hl9XHhe_tXF1x4V-mXjlYw,,&l10n=ru&cts=1512729357277&mc=4.81745484386908
В работе решается задача построения системной модели цифрового изображения, в рамках которой учитывались бы все основные особенности получения цифровых снимков, которые оказывают непосредственное влияние на решение задач сегментации и распознавания изображений.
|
Из за большого объема этот материал размещен на нескольких страницах:
1 2 3 4 5 |
