Kyeong-Hoon Jung

An efficient selection of HOG feature for SVM classification of vehicle

Support Vector Machine (SVM) classifier with Histogram of Oriented Gradients (HOG) feature become one of the most popular techniques used for vehicle detection in recent years. And the computing time of SVM is a main obstacle to get real time implementation which is important for Advanced Driver Assistance Systems (ADAS) applications. One of the effective ways to reduce the computing complexity of SVM is to reduce the dimension of HOG feature. In this paper, we examine the effect of the number of HOG bins on the vehicle detection and the symmetric characteristics of HOG feature of vehicle. And we successfully demonstrate the speed-up of SVM classifier for vehicle detection by about three times while maintaining the detection performance.

Development of partnership between industry and university via customized field-oriented curriculum

This paper introduces the education-oriented collaboration model between LG Electronics and Handong Global University in the form of custom-tailored curriculum - embedded multimedia system track. A new curriculum was developed based on an analysis of the current curriculum with the survey results of industry demands for the engineering graduates, LGEs business direction and comprehensive on-site feedback. The previous curriculum was reshaped by revision of the course contents to incorporate the new track and addition of the custom-tailored specialized courses to suit the needs of LGE. Various programs such as internships and industry scholarships were provided to promote the effectiveness of collaboration. It was mutually recognized that this collaboration model is a good example of a win-win strategy between university and industry.

Spectral characteristics of caries-related autofluorescence spectra and their use for diagnosis of caries stage.

Abstract. The purpose of the present study was to identify factors useful for diagnosis of the caries stage from laser-induced autofluorescence (AF) spectra. Affected teeth were accurately staged and allocated to four groups: sound, stage II, stage III, or stage IV. A 405-nm laser was used to produce AF spectra. The spectrum factors analyzed were spectrum slope at 550 to 600 nm, spectral area from 500 and 590 nm, and intensity ratio of peaks 625 and 667 nm (625/667  nm). DIAGNOdent was used as control measurement. AF spectra of sound teeth had a peak near 500 nm followed by a smooth decline to 800 nm. As caries progressed, some specimens in stages II to IV showed one or two peak(s) near 625 and 667 nm. Slopes at 550 to 600 nm and areas under the curve at 500 to 590 nm were significantly different (p<0.001) for each stage. Two-peak ratios were also significantly different (p<0.001) except for stage III and stage IV. DIAGNOdent readings for sound and stage II and stage III and IV were not significantly different. Among the studied factors, the spectrum slope at 550 to 600 nm and area under curve at 500 to 590 nm could be useful treatment decision-making tools for carious lesions.

Design and analysis of low power image filters toward defect-resilient embedded memories for multimedia socs

In the foreseeable future, System-on-Chip design will suffer from the problem of low yield especially in embedded memories. This can be a critical problem in a multimedia application like H.264 since it needs a huge amount of embedded memory. Existing approaches to solve this problem are not feasible given the higher memory defect density rates in technologies below 90 nm. In this paper, we present a new defect-resilience technique which employs the directional image filter in order to recover data from corrupted embedded memory. According to the analysis based on simulation the proposed filter can greatly improve the visual quality of the defected H.264 video streams with errors in data memory reaching up to 1.0% memory BER (Bit Error Rate) with lower power consumption relative to conventional median filter. Therefore, the proposed method can be a good solution to overcome the problem of low yield in multimedia SoC memory without suffering from additional redundant memory overhead.

Determination of moving direction by the pose transition of vehicle in blind spot area

Vision-based blind spot monitoring algorithm which uses the rear-view dashboard camera has been recently proposed. In order to determine whether the detected vehicle is approaching or not, it needs a motion estimation module. In this paper, we examine the transition of vehicles pose and employ a multi-class HOG and SVM. The relative moving direction can be estimated without use of motion estimation. The experimental results show that the complexity can be considerably reduced with only a slight decrease of recall rate.

A Stereoscopic 3-D Broadcasting System Using Fixed and Mobile Hybrid Delivery and the Quality Assessment of the Mixed Resolution Stereoscopic Video

In this paper, we demonstrate the structure and characteristics of the bandwidth-efficient stereoscopic 3-D broadcasting system entitled service compatible 3-D-TV using main and mobile hybrid delivery (SC-MMH), which enables the broadcasters to provide three different types of services (2-D fixed, mobile, and 3-D) by transmitting only the two legacy broadcast streams (2-D fixed and mobile). By combining the two existing broadcast networks, the proposed system provides high quality 3-D services, fully guarantees backward-compatibility with the legacy fixed/mobile 2-D receivers, and maximizes the channel efficiency. One major feature of the proposed system is bit-rate saving based on the video resolution asymmetry between the left and right eye views, also referred to as “mixed-resolution.” Although it is well-known wisdom that the mixed-resolution stereoscopic videos are perceived close to the higher resolution based on the binocular rivalry suppression theory, the perceived quality also depends on the factors such as scale ratio, viewing distance, bit-rate distribution, and types of codecs. Therefore, we conducted the objective and subjective tests to measure the service quality of the proposed system and find the optimum bit-rate distribution points, and with the results, we demonstrate that SC-MMH can provide the satisfactory quality when the proper level of bit-rate is allocated to the mobile service.

Visual quality improvement for hybrid 3DTV with mixed resolution using conditional replenishment algorithm

This paper proposes the conditional replenishment algorithm (CRA) to improve the visual quality of hybrid stereoscopic 3DTV where the spatial resolutions of left and right views are mismatched. The basic concept of CRA is to determine the better replacement from disparity compensated view and simply enlarged view in generating the enhanced right view. The simulation results show that the proposed CRA can successfully improve the objective quality of the poor view and give positive effect on the final 3D visual quality.

Quality control of conditional replenishment algorithm for hybrid 3DTV with mixed resolution

Conditional replenishment algorithm (CRA) was proposed to improve the visual quality of 3DTV system with mixed resolution. Since CRA generates additional information which is transmitted to a decoder, the bit rate or quality gain needs to be carefully managed. This paper proposes a method to control the quality of images enhanced by CRA as uniform as possible while maintaining the average bit rate. The simulation result shows that the quality can be successfully controlled.

Measurement of Flickering Artifact for H.264 with Periodic I-Frame Structure

Most of multimedia video coding algorithms are lossy schemes and several kinds of spatial and temporal artifacts are inevitable. Flickering, which is the most typical coding artifact in time domain, is mainly due to fact that the quality of coded sequence fluctuates as the quantization parameter is adjusted for rate control. In this paper, we analyzed the effect of quality variation according to the characteristics of video sequence when the I-frames are periodically inserted. And we proposed the FR(Full Reference)-based assessment algorithm to measure the amount of flickering artifacts in the coded video. It is discovered that the flickering becomes critical when the level of quality is intermediate and is affected by the amount of detail or movement, the size of object, and camera parameters. The proposed measurement algorithm shows is well consistent with HVS(Human Visual System).

No-reference Perceptual Quality Assessment of Digital Image

In this paper, a no-reference perceptual metric is proposed for image quality assessment. It measures the amount of overall blockiness and blurring of the image and evaluates the amount of ringing, staircase, and mosaic noises around the strong edges. Finally, the individual scores are combined by a fuzzy integral to generate the final quality score of the image. The quality scores obtained by the proposed algorithm show strong relationship with the MOS(Mean Opinion Score) values by experts.