• Journal of Korea Multimedia Society
• /
• v.14 no.5
• /
• pp.650-657
• /
• 2011

This paper proposes a novel bit allocation algorithm for H.264 scalable extension(SE) based on a human visual system (HVS) to improve the coding efficiency. The proposed algorithm is consist of two stages: visual weighting model and visual weighting-based bit allocation algorithm. In the first stage, the visual weighting for each macroblock (MB) is analyzed according to the region of interests. Then the adaptation of the visual weighting into the bit allocation routine for each quality layer is performed for improving the visual quality. In the simulation results, it is observed that the proposed scheme can improve the subjective and objective video quality in the same bit rate, compared to the previous scalable video coding in H.264.

• Phonetics and Speech Sciences
• /
• v.3 no.2
• /
• pp.65-70
• /
• 2011

Recently, audio-visual speech recognition (AVSR) has been studied to cope with noise problems in speech recognition. In this paper we propose a novel method of deciding weighting factors for audio-visual information fusion. We adopt the particle swarm optimization (PSO) to weighting factor determination. The AVSR experiments show that PSO-based normalized confidence measures (NCM) improve the rejection performance of mis-recognized words by 33%.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.1
• /
• pp.364-380
• /
• 2016

The most popular approach in object classification is based on the bag of visual-words model, which has several fundamental problems that restricting the performance of this method, such as low time efficiency, the synonym and polysemy of visual words, and the lack of spatial information between visual words. In view of this, an object classification based on weakly supervised E2LSH and saliency map weighting is proposed. Firstly, E2LSH (Exact Euclidean Locality Sensitive Hashing) is employed to generate a group of weakly randomized visual dictionary by clustering SIFT features of the training dataset, and the selecting process of hash functions is effectively supervised inspired by the random forest ideas to reduce the randomcity of E2LSH. Secondly, graph-based visual saliency (GBVS) algorithm is applied to detect the saliency map of different images and weight the visual words according to the saliency prior. Finally, saliency map weighted visual language model is carried out to accomplish object classification. Experimental results datasets of Pascal 2007 and Caltech-256 indicate that the distinguishability of objects is effectively improved and our method is superior to the state-of-the-art object classification methods.

• Journal of Korean Society of Water and Wastewater
• /
• v.16 no.6
• /
• pp.686-699
• /
• 2002

The purpose of this study is to estimate deterioration factors and weighting factors in pipe network which each local self-governments takes rehabilitation and replacement work present time. Deterioration factors in pipe network are able to effected of specific province or location related with water supply. Most of water supply pipes are laid under the ground, it is hard to quantify deterioration degree of water system. Moreover, the timing and economic limitation and insufficient information on the spot survey gives a difficulty to look over how old water supply system is. Accordingly, this study collects and analyses five data as the laying environment, visual analysis, analysis of soil contents, analysis of pipe material, and questionary survey data in water pipe of A city. The deterioration factor estimates 14 factors with excavation and experimental analysis and 9 factors without excavation and experimental analysis. Also, the weighting factors are estimated by using the multiple linear regressions and the linear programming. The estimated deterioration factor and weighting results are compared the analysis result of visual, pipe material, and soil contents with the Probabilistic Neural Network Model. Consequently, the model results of estimated 9 factors in this study and 14 factors show the 1-2% difference. The result show that the proposed model could be used to decide the deterioration condition of pipe line with real excavation and experimental analysis.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.12C
• /
• pp.1214-1221
• /
• 2005

In this paper, we propose a digital watermarking method for still images, in which the human visual system (HVS) is used in the discrete cosine transform (DCT) domain. The modulation transfer function (MTF) of the HVS model is employed to increase the invisibility of the inserted watermark in images. The proposed watermarking method is shown to be robust to several common image processing techniques, including lowpass filtering and cropping. Also, using the energy relationship of the DCT, we derive the equation that directly computes the watermark weighting factor in the DCT domain for the specified peak signal to noise ratio (PSNR) of the still image and the length of watermark to be inserted. The difference between desired PSNR and PSNR in spatial domain is within 0.07dB for the 7 test images.

• Journal of Broadcast Engineering
• /
• v.24 no.5
• /
• pp.827-835
• /
• 2019

In the pooling stage of a full reference image quality assessment (FR-IQA) technique, the global perceived quality for any distorted image is usually measured from the quality of its local image patches. But all the image patches do not have equal contribution when estimating the overall visual quality since the degree of degradation on those patches depends on various considerations i.e., types of the patches, types of the distortions, distortion sensitivities of the patches, saliency score of the patches, etc. As a result, weighted pooling strategy comes into account and different weighting mechanisms are used by the existing FR-IQA methods. This paper performs a thorough analysis and proposes a novel weighting function by considering the luminance adaptation as well as the visual saliency effect to offer more appropriate local weights, which can be adopted in the existing FR-IQA frameworks to improve their prediction accuracy. The extended experimental results show the effectiveness of the proposed weighting function.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.18 no.5
• /
• pp.685-694
• /
• 1993

A progressive image transmission (PIT) scheme based on the classified transform vector quantization (CVQ) technique using the lapped orthogonal transform (LOT) and human visual system (HVS) weighting is proposed in this paper. Conventional block transform coding of images using DCT produces in general undesirable block-artifacts at low bit rates. In this paper, image blocks are transformed using the LOT and classified into four classes based on their structural properties and further divided adaptively into subvectors depending on the LOT coefficient statistics with HVS weighting to improve the reconstructed image quality by adaptive bit allocation. The subvectors are vector quantized and transmitted progressively. Coding tests using computer simulations show that the LOT/CVQ based PIT of images is a effective coding scheme. The results are also compared with those obtained using PIT/DCTVQ. The LOT/CVQ based PIT reduces the block-artifacts significantly.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.29B no.11
• /
• pp.93-101
• /
• 1992

HVS based transform coding techniques for still images take the lowest spatial frequency and luminance contrast sensitivity as the most crucial factors in determining the image quality perceived by human eye. Only the frequency weighting has been main issue. The proposed Weighted Discrete Cosine Transform (WDCT) has the lowest frequency of about 4 [cyc/deg] in 8$\times$8 subblock with the contrast-weighted DC coefficient. The error criteria have been rederived to adapt the contrast weighting for performance evaluation. The proposed scheme has produced higher quality images compared to the unweighted and other schemes.

• The Journal of the Acoustical Society of Korea
• /
• v.18 no.4
• /
• pp.78-83
• /
• 1999

This paper concerns a least square method for pattern optimization of a thinned sensor array in which the squared error between a desired pattern and a synthesized one is minimized. A weighting function is applied in the function with respect to the array visual range for a symmetric and asymmetric configuration for sensor spacing. An exponential weighting function is proposed to control the sidelobes efficiently around the mainbeam and to generate a uniform sidelobe. The resulting pattern may be employed to eliminate incoming interferences distributed uniformly around the array visual range.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.40 no.6
• /
• pp.124-131
• /
• 2003

In this paper, we propose a Progressive scalable video coding algorithm using frequency weighting in the DCT domain. Since the human visual system (HVS) can be modeled as a nonlinear point transformation, called the modulation transfer function (MTF), we tan use the frequency weighting matrix to enhance the video image quality. We change this frequency weighting matrix into the frequency shift matrix to apply to the bit-plane coding method for the fine granular scalable (FGS) video coding We also define a new error metric JNDE (just noticeable difference) to measure the perceptual image quality in terms of human vision.