IEEE Transactions on Pattern Analysis and Machine Intelligence

A novel tunnel-lining crack recognition system based on digital image technology

Summary from only given. In this paper, cost minimization of a stand-alone hybrid energy system including photovoltaics, wind turbines and fuel cells through the overall 20-years life time of the system is performed considering reliability constraints. The cost function of the system includes investment cost, operation and maintenance cost and the cost associated with loss of load. The applied wind speed data and solar irradiation data belongs to a region in north west of Iran. To handle the mixed integer nonlinear optimization problem, differential evolutionary algorithm is applied. To provide a time efficient solution process for the optimization problem, an approximated reliability model is used for reliability assessment. Numerical results depict that while the overall system cost is optimized, the reliability indices are within a satisfactory bound with regard to the reliability standards. Comparative results with particle swarm optimization (PSO) present the efficacy of the proposed algorithm.

Economic and reliable design of a hybrid PV-wind-fuel cell energy system using differential evolutionary algorithm

An improved set partition in hierarchical trees (SPIHT) algorithm based on wavelet coefficient blocks is presented for image coding. According to the algorithm characteristic of traditional SPIHT that scanning once can process four wavelet coefficients, our approach is to divide the wavelet transform coefficient matrix into corresponding coefficient blocks and encodes every block with different strategy. When the maximum value of the coefficient block is less than the current threshold, only one bit is used for coding, thus to decrease the number of comparison and reduce the loss of coding efficiency caused by repeat coding and early coding for non-important coefficients. Analysis and simulation results show that the proposed algorithm can provide a better peak signal to noise ratio performance at low bit rate.

An improved SPIHT algorithm based on wavelet coefficient blocks for image coding

Digital image and video image require large amount of capacity to store. Image compression means reducing the size of image, without losing its quality. Depending on the reconstructed image, to be exactly same as the original or some unidentified loss may be incurred, two techniques for compression exist. Two techniques are: lossy techniques and lossless techniques. Digital imaging play an important role in image processing therefore it is necessary to develop a system that produces high degree of compression while preserving critical image. In this paper we present hybrid model which is the combination of several compression techniques. This paper presents DWT and DCT implementation because these are the lossy techniques and also introduce Huffman encoding technique which is lossless. On several medical, Lena images simulation has been carried out. At last implement lossless technique so our PSNR and MSE will go better, and due to DWT and DCT we will get good level of compression without losing any image. When compared to standalone DCT and DWT algorithms,the proposed hybrid algorithm gives better result in term of peak-signal-tonoise ratio with a higher compression ratio.

Comparison and improvement of image compression using dct, dwt & huffman encoding techniques

In order to get a balance of compression quality, compression ratio and time efficiency, we exploit JPEG_LS prediction and context-based encoding, and propose another linear prediction and run length code model. And we give an innovative compression algorithm design to evaluate the performance of our proposed schema. We present the compression results using a large number of images compared with JPEG_LS. As a result, our proposed schema obtains the same compression quality, an increasing compression speed as much as 150%–250%, at the expense of compression ratio decreasing of 20%–35%. Also our proposed algorithm has less computational complexity. So it can be included that our proposed algorithm is more suitable for real-time and embedded applications than JPEG-LS.

A fast and low complicated image compression algorithm for predictor of JPEG-LS

A critical and Indispensable step of an automatic fingerprint identification system (AFIS) is Fingerprint image Thinning. many algorithms have been devised and applied trying to obtain one pixel wide and 8 neighbor connected thinning result. In this paper, a method of improving the thinning result through adding eliminate structure templates is proposed. The ZS algorithm is a parallel thinning algorithm which has fast converge speed, but it also has some problems such as Distortion of the fingerprint ridge, incapability of gaining one-pixel wide skeleton. the proposed improved algorithm will solve the problems without paying extra times, as a result, more effective fingerprint minutiae points could be extracted.

An approach to improve the ZS thinning algorithm

Motivated by the reported out performance in the fingerprint recognition thesises of PCA by ICA in the linear case where only a single factor is allowed to vary, and the outperformance of PCA by FET when multiple factors are allowed to vary, it is natural to ask whether there a multilinear generalization of ICA and if its performance is better than the other two methods. In this paper, We have presented a multilinear algebraic framework for fingerprint image recognition, which employs a tensor (N-mode) extension of the conventional matrix SVD. We also introduced a multilinear projection algorithm for fingerprint recognition, which projects an unlabeled test image into the N constituent mode spaces to infer its mode label-sperson, finger, type

A novel approach of fingerprint recognition based on multilinear ICA

By analyzing the characteristics of degree zerotree, both the EZW wavelet coder and SPIHT wavelet coder have advantages and disadvantages. Combined with the characteristics of wavelet coefficients, a zerotree based on hybrid encoder program was found. With the wavelet zerotree definition, taking the pre-processing zerotree before the zerotree coding, the identification of suitable programs of zerotree encoder comes out. The simulation results show that comparing with the separate use of EZW or SPIHT encoder, the hybrid encoder's encoding algorithm can effectively enhance the efficiency of code compression PSNR.

Xiaojun Jing

Papers

A fast and low complicated image compression algorithm for predictor of JPEG-LS

An approach to improve the ZS thinning algorithm

A novel approach of fingerprint recognition based on multilinear ICA

Improved zero-tree coder research based on wavelet coefficients