• The Journal of the Korea Contents Association
• /
• v.14 no.12
• /
• pp.555-564
• /
• 2014

The boundary noise in image syntheses using DIBR consists of noisy pixels that are separated from foreground objects into background region. It is generated mainly by edge misalignment between the reference image and depth map or blurred edge in the reference image. Since hole areas are generally filled with neighboring pixels, boundary noise adjacent to the hole is the main cause of quality degradation in synthesized images. To solve this problem, a new boundary noise removal algorithm using a preprocessing of the depth map is proposed in this paper. The most common way to eliminate boundary noise caused by boundary misalignment is to modify depth map so that the boundary of the depth map can be matched to that of the reference image. Most conventional methods, however, show poor performances of boundary detection especially in blurred edge, because they are based on a simple boundary search algorithm which exploits signal gradient. In the proposed method, a two-step hierarchical approach for boundary detection is adopted which enables effective boundary detection between the transition and background regions. Experimental results show that the proposed method outperforms conventional ones subjectively and objectively.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.21 no.8
• /
• pp.87-95
• /
• 2007

In this paper, we analyzed the mr detection and the stability of the object tracking system by an adaptive stereo object hacking using region-based MAD(Mean Absolute Difference) algorithm and the modified PID(Proportional Integral Derivative)-based pan/tilt controller. That is, in the proposed system, the location coordinates of the target object in the right and left images are extracted from the sequential stereo input image by applying a region-based MAD algorithm and the configuration parameter of the stereo camera, and then these values could effectively control to pan/tilt of the stereo camera under the noisy circumstances through the modified PID controller. Accordingly, an adaptive control effect of a moving object can be analyzed through the advanced system with the proposed 3D robot vision, in which the possibility of real-time implementation of the robot vision system is also confirmed.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.4
• /
• pp.34-40
• /
• 2018

This paper presents a new method to estimate the pose of a moving object accurately using a monocular camera and a low-end GPS+IMU sensor system. For this goal, we adopted a deep neural network for the semantic segmentation of input images and compared the results with a semantic map of a neighborhood. In this map matching, we use weight tables to deal with label inconsistency effectively. Signals from a low-end GPS+IMU sensor system are used to limit search spaces and minimize the proposed function. For the evaluation, we added noise to the signals from a high-end GPS-IMU system. The results show that the pose can be recovered from the noisy signals. We also show that the proposed method is effective in handling non-open-sky situations.

• Journal of Computational Design and Engineering
• /
• v.1 no.4
• /
• pp.272-288
• /
• 2014

This paper presents a two-step, semi-automated method for reconstructing a three-dimensional (3D) shape of the prostate from a 3D transrectal ultrasound (TRUS) image. While the method has been developed for prostate ultrasound imaging, it can potentially be applicable to any other organ of the body and other imaging modalities. The proposed method takes as input a 3D TRUS image and generates a watertight 3D surface model of the prostate. In the first step, the system lets the user visualize and navigate through the input volumetric image by displaying cross sectional views oriented in arbitrary directions. The user then draws partial/full contours on selected cross sectional views. In the second step, the method automatically generates a watertight 3D surface of the prostate by fitting a deformable spherical template to the set of user-specified contours. Since the method allows the user to select the best cross-sectional directions and draw only clearly recognizable partial or full contours, the user can avoid time-consuming and inaccurate guesswork on where prostate contours are located. By avoiding the usage of noisy, incomprehensible portions of the TRUS image, the proposed method yields more accurate prostate shapes than conventional methods that demand complete cross-sectional contours selected manually, or automatically using an image processing tool. Our experiments confirmed that a 3D watertight surface of the prostate can be generated within five minutes even from a volumetric image with a high level of speckles and shadow noises.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.18 no.11
• /
• pp.1635-1644
• /
• 1993

In this paper, we examine the statistical characteristics of edge enhancing filters and their efficacy as preprocessing operator before edge detection. In particular, edge enhancing filters called the Comparison and Selection(CS), Hachimura-kuwahara(HK), and Selective Average(SA) filters are considered. These filters can reduce noise while producing step-type edges, thus seem to be effective for preprocessing noisy images prior to applying edge detecors. The ability of edge enhancing filters to suppress white Gaussian noise and the error probabilities occured during the edge detection following SA prefiltering are evaluated statistically through numerical analysis. The effect of prefiltering on edge detection is assessed by applying the edge enhancing fitters to a noise image degraded by additive white noise prior to applying the Sobel operator and the Laplacian of Gaussian( LoG ) operator, respectively. It is shown that the edge enhancing filters tend to produce ideal step-type edges while reducing the noise reasonably well, and the use of edge enhancing filters prior to edge detection can improve the performance of subsequent edge detector.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.37 no.6
• /
• pp.10-21
• /
• 2000

The Cellular Noninear Networks structure for Distance Transform(DT) and the robust optical flow detection algorithm based on the DT are proposed. For some applications of optical flows such as target tracking and camera ego-motion computation, correct optical flows at a few feature points are more useful than unreliable one at every pixel point. The proposed algorithm is for detecting the optical flows on the trajectories only of the feature points. The translation lengths and the directions of feature movements are detected on the trajectories of feature points on which Distance Transform Field is developed. The robustness caused from the use of the Distance Transform and the easiness of hardware implementation with local analog circuits are the properties of the proposed structure. To verify the performance of the proposed structure and the algorithm, simulation has been done about various images under different noisy environment.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.8
• /
• pp.97-105
• /
• 2015

Content-adaptive steganography methods embed secret messages in hard-to-model regions of covers such as complicated texture or noisy area. Content-adaptive steganalysis methods often need high dimensional features to capture more subtle relationships of local dependencies among adjacent pixels. However, these methods require many computational complexity and depend on the location of hidden message and the exploited distortion metrics. In this paper, we propose an improved steganalysis method for content-adaptive steganography to enhance detection rate with small number features. We first show that the features form the difference between DCT coefficients are useful for analyzing the content-adaptive steganography methods, and present feature extraction mehtod using first-order Markov probability for the the difference between DCT coefficients. The extracted features are used as input of ensemble classifier. Experimental results show that the proposed method outperforms previous schemes in terms of detection rates and accuracy in spite of a small number features in various content-adaptive stego images.

• Journal of Broadcast Engineering
• /
• v.1 no.1
• /
• pp.7-13
• /
• 1996

MASF is a kind of temporal filter proposed for noise reduction and temporal band limitation. MASF uses motion vectors to extract temporal information in spatial domain. Therefore, inaccurate motion information causes some distortions in MASF operation. Currently, bilinear interpolation after BMA(Block Matching Algorithm) is used for the motion estimation sheme of MASF. But, this method results in unreliable estimation when the object in image sequence has larger movement than the maximum displacement assumed in BMA or the input images are severely corrupted with noise. In order to i:;olve this problem, we analyse the effect of inaccurate motion on MASF and propose a hierarchical motion estimation algorithm based on the analysis results. Experimental results show that the proposed method produces reliable output under large motion and noisy situations.

• Transactions of the Society of Information Storage Systems
• /
• v.1 no.2
• /
• pp.155-160
• /
• 2005

In this paper, the channel decoder promising reliable data retrieving in noisy holographic channel has been developed for holographic WORM(write once read many) system. It covers various DSP(digital signal processing) blocks, such as align mark detector, adaptive channel equalizer, modulation decoder and ECC(error correction code) decoder. The specific schemes of DSP are designed to reduce the effect of noises in holographic WORM(H-WORM) system, particularly in prototype of DAEWOO electronics(DEPROTO). For real time data retrieving, the channel decoder is redesigned for FPGA(field programmable gate array) based hardware, where DSP blocks calculate in parallel sense with memory buffers between blocks and controllers for driving peripherals of FPGA. As an input source of the experiments, MPEG2 TS(transport stream) data was used and recorded to DEPROTO system. During retrieving, the CCD(charge coupled device), capturing device of DEPROTO, detects retrieved images and transmits signals of them to the FPGA of hardware channel decoder. Finally, the output data stream of the channel decoder was transferred to the MPEG decoding board for monitoring video signals. The experimental results showed the error corrected BER(bit error rate) of less than $10^{-9}$, from the raw BER of DEPROTO, about $10^{-3}$. With the developed hardware channel decoder, the real-time video demonstration was possible during the experiments. The operating clock of the FPGA was 60 MHz, of which speed was capable of decoding up to 120 mega channel bits per sec.

• Journal of the Institute of Convergence Signal Processing
• /
• v.1 no.1
• /
• pp.23-31
• /
• 2000

Iterative curve evolution techniques are powerful methods for image segmentation. Classical methods proposed curve evolutions which guarantee close contours at convergence and, combined with the level set method, they easily handled curve topology changes. In this paper, we present a new geometric active contour model based on level set methods introduced by Osher & Sethian for detection of object boundaries or shape and we adopt anisotropic diffusion filtering method for removing noise from original image. Classical methods allow only one-way curve evolutions : shrinking or expanding of the curve. Thus, the initial curve must encircle all the objects to be segmented or several curves must be used, each one totally inside one object. But our method allows a two-way curve evolution : parts of the curve evolve in the outward direction while others evolve in the inward direction. It offers much more freedom in the initial curve position than with a classical geodesic search method. Our algorithm performs accurate and precise segmentations from noisy images with complex objects(jncluding sharp angles, deep concavities or holes), Besides it easily handled curve topology changes. In order to minimize the processing time, we use the narrow band method which allows us to perform calculations in the neighborhood of the contour and not in the whole image.