• ETRI Journal
• /
• v.31 no.1
• /
• pp.1-9
• /
• 2009

Frequency domain analysis is a fundamental procedure for understanding the characteristics of visual data. Several studies have been conducted with 2D videos, but analysis of stereoscopic 3D videos is rarely carried out. In this paper, we derive the Fourier transform of a simplified 3D video signal and analyze how a 3D video is influenced by disparity and motion in terms of temporal aliasing. It is already known that object motion affects temporal frequency characteristics of a time-varying image sequence. In our analysis, we show that a 3D video is influenced not only by motion but also by disparity. Based on this conclusion, we present a temporal anti-aliasing filter for a 3D video. Since the human process of depth perception mainly determines the quality of a reproduced 3D image, 2D image processing techniques are not directly applicable to 3D images. The analysis presented in this paper will be useful for reducing undesirable visual artifacts in 3D video as well as for assisting the development of relevant technologies.

• Proceedings of the IEEK Conference
• /
• 2001.06d
• /
• pp.53-56
• /
• 2001

This paper presents a computationally efficient post-processing algorithm for HDTV. The proposed algorithm can reduce both blocking artifacts and mosquito noise while preserving the sharpness and naturalness of the reconstructed video signal. Performance improvements compared with other techniques are obtained according to simulation results.

• Journal of the Institute of Convergence Signal Processing
• /
• v.6 no.4
• /
• pp.169-175
• /
• 2005

In recent, implementation of the embedded system is gradually in the spotlight of world-wide by information technology(IT) engineers. By this time, an implementation of real time system is limited on image acquisition and processing system in practical. In this paper, the USB camera interface system based on the embedded linux OS is implemented using USB 2.0 camera with low cost. This system can obtain image signals into the memory via X-hyper255B processor from USB camera. It is need to initialize USB camera by the Video4Linux for the kernel device driver. From the system image capturing and image processing can be performed. It is confirmed that the image data can be transformed to packet of Network File System(NFS) and connected to the internetwork, then the data can be monitored from the client computer connected to the internetwork.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.17 no.5
• /
• pp.529-538
• /
• 1992

Recently, image signal processing techniques for HDTV signal have been drastically developed. This kind of skill improvement on signal processing need specific memory device for video signal. in this paper, data latch scheme which implements CMOS flip-flop to hold Information from in-put strobe and new reading method is devised to attain a proper access time suitable for HDTY signal. Compared with conventional write scheme, data latch method has two procedures to complete write operation : bit line write and storage cell write, enabling concurrent I /0 operation at the same address. Also, fast read access is possible through the method similar to static column mode and the separated read word line.

• Journal of the Korea Society of Computer and Information
• /
• v.15 no.3
• /
• pp.37-43
• /
• 2010

In this study, the multi format 8.4" LCD monitor for broadcasting has been proposed by the simple circuit design and the efficient processing technique. The small sizing technique, the mixed-type video signal processing technique, the emerging skill of multi-format HD signal to one port, the dividing technique between the analog and digital signal, the embedded audio signal developing technique are the key researches of this study. The proposed multi format 8.4" LCD monitor has been focused on the broadcasting field and has been supposed to have the efficiency at that field.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.6 no.2
• /
• pp.206-211
• /
• 2011

In this paper, a fire detection algorithm based on video processing is proposed. At the first stage, background image extracted from CCTV video input signal, and then foreground image were separated by differencing CCTV input signal from background image. At the second stage, candidated area were extracted by using color information from foreground image. At the final stage, smoke or flame characteristic area were separated by using Gaussian mixture modeling applied to candidated area, and then fire can be detected. Through real experiments at the inner room, it is shown that the proposed system works well.

• Journal of Korea Multimedia Society
• /
• v.25 no.5
• /
• pp.676-684
• /
• 2022

As realistic media are widespread in various image processing areas, image or video compression is one of the key technologies to enable real-time applications with limited network bandwidth. Generally, image or video compression cause the unnecessary compression artifacts, such as blocking artifacts and ringing effects. In this study, we propose a Deep Residual Channel-attention Network, so called DRCAN, which consists of an input layer, a feature extractor and an output layer. Experimental results showed that the proposed DRCAN can reduced the total memory size and the inference time by as low as 47% and 59%, respectively. In addition, DRCAN can achieve a better peak signal-to-noise ratio and structural similarity index measure for compressed images compared to the previous methods.

• Journal of Broadcast Engineering
• /
• v.17 no.2
• /
• pp.223-236
• /
• 2012

We develop a video processing method that allows the more advanced human perception oriented video coding. The proposed method necessarily reflects all influences by the rate-distortion based optimization and the human visual perception that is affected by the visual saliency, the limited space-time resolution and the regional moving history. For reflecting the human perceptual effects, we devise an online moving pattern classifier using the Hedge algorithm. Then, we embed the existing visual saliency into the proposed moving patterns so as to establish a human visual perception model. In order to realize the proposed human visual perception model, we extend the conventional foveation filtering method. Compared to the conventional foveation filter only smoothing less stimulus video signals, the developed foveation filter can locally smooth and enhance signals according to the human visual perception without causing any artifacts. Due to signal enhancement, the developed foveation filter more efficiently transfers the bandwidth saved at smoothed signals to the enhanced signals. Performance evaluation verifies that the proposed video processing method satisfies the overall video quality, while improving the perceptual quality by 12%~44%.

• Journal of Satellite, Information and Communications
• /
• v.11 no.4
• /
• pp.1-9
• /
• 2016

This paper was studied in two ways with respect to the information received from the video signal separation technique of PPI Scop radar device. The proposed technique consists in generating an image signal through the video signal separation and synthesis, symbol generation, the residual image signal generation process. This technology can greatly improve the operating convenience with improved ease of discrimination, screen readability for the operator in analyzing radar information. The first proposed method was constructed for high-speed FPGA-based information processing systems for high speed operation stability of the system. The second proposed method was implemented intelligent algorithms and a software algorithm function curve associated resources.This was required to meet the constraints on the radar information, analysis system. Existing radar systems have not the frame data analysis unit image. However, this study was designed to image data stored in the frame-by-frame analysis of radar images with express information MPEG4 video. Key research content is to highlight the key observations expresses the target, the object-specific monitoring information to the positive image processing algorithm and the function curve delays. For high-definition video, high-speed to implement data analysis and expressing a variety of information was applied to the ARM Processor Support in Pro ASIC3.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.18 no.11
• /
• pp.1645-1654
• /
• 1993

Unlike a noise removal recursive temporal filter, this paper presents a temporal filter which improves visual quality and data compression efficiency. In general, for the temporal band-limitation, temporal aliasing should be considered. Since most of a video signal has temporally aliased components, it is desirable to consider them. From a signal processing point of view, it is impossible to realize the filtering not afeced by the aliasings. However, in this paper, efficient filtering with de-aliasing characteristics is proposed. Considering the location of a video signal, temporal filtering can be accomplished by the spatial filtering along the motion vector trajectory (Motion Adaptive Spatial Filter). This filtered result dose not include the aliasings. Besides the efficient band-limitation, temporal noise is also reduced. For the evaluation of the MASF, its realization and filtering characteristics will be discussed in ditail.