Abstract:In image steganalysis applications, the urgent problem is how todesign an effective classification architecture for given large-scalefeatures. In this paper we propose an approach toautomatically improve the current classification rate under the specific steganographic methods base on ELM feature selection. This approach referred asELM revisited feature selection (ELM-RFS) can tune input weights interm of minimizing mean error. This idea is derived from cross-validation learning and one-dimensional optimal search. During updating of inputweights, we seek the energy decreasing direction directly to discard the influence of leastsignificant features using the leave-one-out (LOO) selection, instead of the updated one-dimensional search at the current energy decreasing direction. The experimental resultsdemonstrate and verify that the new approach is less classification errorthan other classifier such as SVM, direct ELM-LOO learning andconventional ELM in state-of-the-art features.

Alert Name:
Alerting to:
Authentication email will be sent to your email address in 24 hours
Frequency:
Email Message Format:	Plain text Graphical(HTML)

Complete the form below and we will recommend the selected titles to your friends on your behalf. * Indicates a required field.
Your name*:
Your email address*:
Recipient's name*:
Recipient's email address*:
(multiple recipient's names and email addresses should be separated with semicolons)
Your comments:	I thought you would find the page(s) useful.

Your name:
Your email address:
Recipient's name:
Recipient's email address:
(multiple recipient's names and email addresses should be separated with semicolons)
Your comments:	I thought you would find this page useful.

Disclaimer: This message was sent to your friend using the "Send it to a friend" facility on the Sciencepaper Online’ WWW site, http://www.paper.edu.cn/en. The Sciencepaper Online is not responsible for the content of this email, and anything said in this email does not necessarily reflect the Sciencepaper Online's views.


	1. ELM-based image steganalysis
	FENG Guo-Rui, ZHANG Hai-Yan
	Electrics, Communication and Autocontrol Technology 17 December 2013
	Show/Hide Abstract \| Cite this paper︱Full-text: PDF (0 B)

	Abstract:In image steganalysis applications, the urgent problem is how todesign an effective classification architecture for given large-scalefeatures. In this paper we propose an approach toautomatically improve the current classification rate under the specific steganographic methods base on ELM feature selection. This approach referred asELM revisited feature selection (ELM-RFS) can tune input weights interm of minimizing mean error. This idea is derived from cross-validation learning and one-dimensional optimal search. During updating of inputweights, we seek the energy decreasing direction directly to discard the influence of leastsignificant features using the leave-one-out (LOO) selection, instead of the updated one-dimensional search at the current energy decreasing direction. The experimental resultsdemonstrate and verify that the new approach is less classification errorthan other classifier such as SVM, direct ELM-LOO learning andconventional ELM in state-of-the-art features.
	TO cite this article:FENG Guo-Rui, ZHANG Hai-Yan. ELM-based image steganalysis[OL].[17 December 2013] http://en.paper.edu.cn/en_releasepaper/content/4569229


	2. EMD-SIFT based Deep Space Image Matching Method
	Yin Mengzheng,Wu Xuechen,Li Xiaoyu,Feng Hongqi,Wang Qiang
	Electrics, Communication and Autocontrol Technology 24 May 2013
	Show/Hide Abstract \| Cite this paper︱Full-text: PDF (0 B)

	Abstract:The paper puts forward a kind of scale invariant feature matching method for asteroid image sequences. In which we apply EMD (Empirical Mode Decomposition) to extract the features from image sequences, SIFT (scale invariant feature transform) to match the extracted features, and affine transform to stretch and rotate the images. We can achieve a global view of the scanned asteroid surface. Experiment results demonstrate the excellent performance of the EMD-SIFT based method in mosaicking the asteroid image sequences.
	TO cite this article:Yin Mengzheng,Wu Xuechen,Li Xiaoyu, et al. EMD-SIFT based Deep Space Image Matching Method[OL].[24 May 2013] http://en.paper.edu.cn/en_releasepaper/content/4544596

Saved Papers

Saved Papers


	3. A Differential Approach for Computationally Efficient Recovery of Radiometric Response Function
	GU Zhenghui
	Electrics, Communication and Autocontrol Technology 28 January 2013
	Show/Hide Abstract \| Cite this paper︱Full-text: PDF (0 B)

	Abstract:In high dynamic range imaging, it is important to estimate the radiometric response function of a camera that relates image pixel values to scene radiance. Among numerous existing approaches, Debevec and Malik's algorithm has been widely used to determine the response function from multiple differently exposed photographs. However, with this algorithm, the computational complexity goes up when the number of exposures decreases. The problem becomes serious in the case of two exposures. In this paper, we present a differential approach that can significantly reduce the complexity of their algorithm at small numbers of exposures. Experimental results illustrate the performance of the proposed method.
	TO cite this article:GU Zhenghui. A Differential Approach for Computationally Efficient Recovery of Radiometric Response Function[OL].[28 January 2013] http://en.paper.edu.cn/en_releasepaper/content/4516884


	4. A correlative two-step approach to hallucinating faces
	Huanxi Liu
	Electrics, Communication and Autocontrol Technology 07 January 2013
	Show/Hide Abstract \| Cite this paper︱Full-text: PDF (0 B)

	Abstract:Face hallucination is to synthesize high-resolution face image from the input low-resolution one. Although many two-step learning-based face hallucination approaches have been developed, they suffer from the expensive computational cost due to the separate calculation of the global and local models. To overcome this problem, we propose a correlative two-step learning-based face hallucination approach which bridges the gap between the global model and the local model. In the global phase, we build a global face hallucination framework by combining the steerable pyramid decomposition and the reconstruction. In the residue compensation phase, based on the combination weights and constituent samples obtained in the global phase, a residue face image is synthesized by the neighbor reconstruction algorithm to compensate the hallucinated global face image with detailed facial features. The ultimate hallucinated face image is the composition of the global face image and the residue face image. Compared with existing approaches, in the global phase, our global face image is more similar to the original high-resolution face image. Moreover, in the residue compensation phase, we use the combination weights and constituent samples obtained in the global phase to compute the residue face image, by which the computational complexity can be greatly reduced without compromising the quality of facial details. The experimental results and comparisons demonstrate that our approach can not only synthesize distinct high-resolution face images efficiently, but also has high computational efficiency.
	TO cite this article:Huanxi Liu. A correlative two-step approach to hallucinating faces[OL].[ 7 January 2013] http://en.paper.edu.cn/en_releasepaper/content/4512944


	5. Joint bit allocation and rate control for multi-view video plus depth based 3D video coding
	Shao Feng
	Electrics, Communication and Autocontrol Technology 11 March 2012
	Show/Hide Abstract \| Cite this paper︱Full-text: PDF (0 B)

	Abstract:In three-dimensional (3D) video coding, the distortions in texture videos and depth maps can all affect the quality of the synthesized virtual view. Therefore, under the total bitrate constraint, effective bit allocation between texture and depth is very important for 3D video coding. In this paper, a joint bit allocation and rate control method for multi-view video plus depth (MVD) based 3D video coding is proposed. We firstly derive a view synthesis distortion model to characterize the effects of coding distortions of texture videos and depth maps on the synthesized virtual view. Based on this model, we derive a rate-distortion model to characterize the relationship between the bitrate and the view synthesis distortion, and the optimal bitrate ratio between texture and depth is established by solving optimization problem. Then, rate control algorithm is performed on view level, texture/depth level and frame level. Experimental results show that compared with other method, the proposed bit allocation method can obtain higher performance of view synthesis. Moreover, the proposed rate control method can accurately control the bitrate to satisfy the total bitrate constraint.
	TO cite this article:Shao Feng. Joint bit allocation and rate control for multi-view video plus depth based 3D video coding[OL].[11 March 2012] http://en.paper.edu.cn/en_releasepaper/content/4471143


	6. Resolution-Scalable Motion Vector Coding Based on Inter-Resolution Prediction
	Li Xiangwei,Lan Xuguang,Zhu ling,Xue Jianru,Zheng Nanning
	Electrics, Communication and Autocontrol Technology 05 January 2012
	Show/Hide Abstract \| Cite this paper︱Full-text: PDF (0 B)

	Abstract:Resolution scalability is an important feature in next generation video coding and appropriate scalable motion information coding technique can improve its overall coding efficiency. This paper presents an efficient resolution-scalable motion vector coding algorithm using inter-resolution based motion vector prediction and scalable multi-resolution motion vector embedded coding technique. The conducted experiments demonstrate significant benefits of the proposed scalable motion vector coding scheme to improve the compression performance of resolution scalability video coding, especially at low resolution.
	TO cite this article:Li Xiangwei,Lan Xuguang,Zhu ling, et al. Resolution-Scalable Motion Vector Coding Based on Inter-Resolution Prediction[OL].[ 5 January 2012] http://en.paper.edu.cn/en_releasepaper/content/4458049


	7. A Novel Watermark Strategy For Quantum Images
	Wei-Wei Zhang,Fei Gao,Bin Liu,Heng-Yue Jia,Qiao-Yan Wen,Hui Chen
	Electrics, Communication and Autocontrol Technology 29 December 2011
	Show/Hide Abstract \| Cite this paper︱Full-text: PDF (0 B)

	Abstract:With the development of quantum computer andquantum network, the security of quantum data becomes moreand more important. Quantum watermarking is the techniquewhich embeds the invisible quantum signal such as the owner’sidentification into quantum multimedia data (such as audio,video and image) for copyright protection. In this paper, a novelwatermark strategy for quantum images is proposed based onflexible representation for quantum images (FRQI). Comparedwith the former strategy (only one based on quantum images)which can only be used to verify the identity of the true ownerof a carrier image, the proposed method can also be used to findout who is the real owner. And it is impossible for anyone exceptthe copyrighter to clear off or extract the watermark images.
	TO cite this article:Wei-Wei Zhang,Fei Gao,Bin Liu, et al. A Novel Watermark Strategy For Quantum Images[OL].[29 December 2011] http://en.paper.edu.cn/en_releasepaper/content/4458478


	8. Fast Trademark Recognition Based on Shape Context and Hierarchical Clustering
	WU Qian,ZHANG Honggang,CHAI Lunshao
	Electrics, Communication and Autocontrol Technology 19 December 2011
	Show/Hide Abstract \| Cite this paper︱Full-text: PDF (0 B)

	Abstract:Trademark recognition has been played a very important part in our daily life, which can be applied to various fields such as advertisement, logistics transportation network, E-commerce and so on. In this paper, a fast trademark recognition method based on shape context is proposed to detect and classify the trademarks in images. However, the difficulties lie in that when the sample set becomes larger, along with the calculation of the feature points matching increases, the computational efficiency greatly reduced. The method proposed in this paper attempts to cluster the feature vectors in training procedure so that the search scope of K nearest neighbors in KNN algorithm could be narrowed. The experimental results show that the modified algorithm has faster computing speed on the basis of maintaining the previous performance, and do well in trademark recognition.
	TO cite this article:WU Qian,ZHANG Honggang,CHAI Lunshao. Fast Trademark Recognition Based on Shape Context and Hierarchical Clustering[OL].[19 December 2011] http://en.paper.edu.cn/en_releasepaper/content/4456540


	9. An image quality metric based on bionic models
	HOU Weilong,GAO Xinbo,HE Lihuo,GAO Fei
	Electrics, Communication and Autocontrol Technology 22 August 2011
	Show/Hide Abstract \| Cite this paper︱Full-text: PDF (0 B)

	Abstract:Image quality evaluation is to use some computational models to predict the quality of the specified image automatically and accurately. Since human eyes are ultimate receptor of images, it is better to mimic human visual system (HVS) to perceive the image quality. Based on the properties of the HVS, a novel bionic image quality metric (IQA) is proposed, which adopts several bionic characteristics, e.g. multi-channel decomposition, contrast sensitivity function, center-surround operation and lateral inhibition mechanism. Experimental results demonstrate that the performance of the proposed IQA method outperforms those of the existing methods.
	TO cite this article:HOU Weilong,GAO Xinbo,HE Lihuo, et al. An image quality metric based on bionic models[OL].[22 August 2011] http://en.paper.edu.cn/en_releasepaper/content/4440162


	10. Improved DCT-domain Color Image Enhancement through Anisotropic Diffusion
	Wan Yi ,Ma Yafeng
	Electrics, Communication and Autocontrol Technology 28 December 2009
	Show/Hide Abstract \| Cite this paper︱Full-text: PDF (0 B)

	Abstract:Color image enhancement directly in the DCT domain is an efficient approach in high speed data transmission. However, such methods often suffer from blocking artifact due to inappropriate scaling between adjacent blocks. We propose an improved method based on anisotropic diffusion that overcomes this difficulty. Simulation results show that the proposed method can effectively remove blocking artifact introduced during the DCT coefficient scaling stage, resulting in much improved performance over some of state of the art methods.
	TO cite this article:Wan Yi ,Ma Yafeng . Improved DCT-domain Color Image Enhancement through Anisotropic Diffusion[OL].[28 December 2009] http://en.paper.edu.cn/en_releasepaper/content/38136

	Check out RSS, or use RSS reader to subscribe this item