• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.11 no.4
• /
• pp.293-298
• /
• 2011

In this paper, we propose a new image descriptor, that is, a rotation invariant histogram of oriented gradients (RIHOG). RIHOG overcomes a disadvantage of the histogram of oriented gradients (HOG), which is very sensitive to image rotation. The HOG only uses magnitude values of a pixel without considering neighboring pixels. The RIHOG uses the accumulated relative magnitude values of corresponding relative orientation calculated with neighboring pixels, which has an effect on reducing the sensitivity to image rotation. The performance of RIHOG is verified via the index of classification and classification of Brodatz texture data.

• Journal of Sensor Science and Technology
• /
• v.27 no.2
• /
• pp.137-141
• /
• 2018

Detecting the ego-lane of a vehicle (the lane on which the vehicle is currently running) is one of the basic techniques for a smart car. Vision sensing is a widely-used method for the ego-lane detection. Existing studies usually find road lane lines by detecting edge pixels in the image from a vehicle camera, and then connecting the edge pixels using Hough Transform. However, this approach takes rather long processing time, and too many straight lines are often detected resulting in false detections in various road conditions. In this paper, we find the lane lines by scanning only a limited number of horizontal lines within a small image region of interest. The horizontal image line scan replaces the edge detection process of existing methods. Automatic thresholding and spatiotemporal filtering procedures are also proposed in order to make our method reliable. In the experiments using real road images of different conditions, the proposed method resulted in high success rate.

• Korean Journal of Remote Sensing
• /
• v.28 no.3
• /
• pp.287-296
• /
• 2012

Although low spatial resolution satellite images like MODIS and GOCI can be important to observe land surface, it is often difficult to visually interpret the imagery because of the low contrast by prevailing cloud covers. We proposed a simple and adaptive stretching algorithm to enhance image contrast over land areas in cloudy images. The proposed method is basically a linear algorithm that stretches only non-cloud pixels. The adaptive linear stretch method uses two values: the low limit (L) from image statistics and upper limit (U) from low boundary value of cloud pixels. The cloud pixel value was automatically determined by pre-developed empirical function for each spectral band. We used MODIS and GOCI images having various types of cloud distributions and coverage. The adaptive contrast stretching method was evaluated by both visual interpretation and statistical distribution of displayed brightness values.

• 한국가시화정보학회:학술대회논문집
• /
• 2005.12a
• /
• pp.36-39
• /
• 2005

Information on temporal evolution of whole velocity fields is essential for physical understanding of a complicated turbulent flow and was obtainable using dynamic PIV because of advances of high-speed imaging technique, laser and electronics. A dynamic PIV systme consists of a high-speed CMOS camera having $1K\times1K$ pixels resolution at 1 KHz and a high-repetition Nd:Yag pulse laser. In order to validate its performance, the dynamic PIV system was applied to a turbulent jet whose Reynolds number is about 3000. The particle images of $1024\times512$ pixels were captured at a sampling rate of 4 KHz. The dynamic PIV system measured successfully the temporal evolution of instantaneous velocity fields of the turbulent jet, from which spectral analysis of turbulent structure was also feasible.

• Proceedings of the Korean Information Science Society Conference
• /
• 2007.06c
• /
• pp.403-406
• /
• 2007

This paper proposes an effective method for text detection in video. First, we apply an edge detection method to the video frame with a relative low threshold to keep all possible text edge pixels. Second, a multi-frame integration method is applied to significantly remove background pixels which are not stationary in a specific period. Finally, text regions are extracted by using the coarse to fine projection method. Experimental results demonstrate the effectiveness of the proposed method.

• IE interfaces
• /
• v.15 no.1
• /
• pp.20-25
• /
• 2002

In this research, a nameplate engraving system for images and texts downloaded through Internet on nameplate is developed. The system consists of two subsystems: thinning algorithm and NC code generation module. In the thinning algorithm, the concept of connectivity is used and center lines of images and texts, which will be used as NC tool paths, can be obtained successfully. Because the center lines are composed of a lot of pixels, NC code would be too long. In the NC code generation module, many useless pixel data are removed and linear interpolation algorithm is applied to only the remaining pixels. By performing actual experiments, the thinning algorithm and the NC code generation module are verified.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.11 no.1
• /
• pp.201-210
• /
• 2003

In this paper, we propose a new binocular stereo correspondence method by maximizing a fitness formulated by integrating two constraints of edge similarity and disparity smoothness simultaneously. The proposed stereopsis focusing to measure distances to leading vehicles on roads uses intensity gradients as matching attribute. In contrast to the previous work of area-based stereo matching, in which matching unit is a pixel, the matching unit of the proposed method becomes an area itself which is obtained by selecting a series of pixels enclosed by two pixels on the left and right boundaries of an object. This approach allows us to cope with real-time processing and to avoid window size selection problems arising from conventional area-based stereo.

• 한국정보디스플레이학회:학술대회논문집
• /
• 2007.08b
• /
• pp.1789-1791
• /
• 2007

We fabricated OTFT-OLED display panel by using Ag-paste for source and drains electrode of OTFTs. The OTFTs were fabricated by solution processes such as spin-coating for PVP gate dielectric and screen printing for S/D electrodes with Ag-paste, except pentacene active layer which was deposited by evaporation. The mobility was 0.024 cm2/V.sec , off state current ${\sim}10-11A$, threshold voltage 7.6 V and on/off current ratio ${\sim}105$. The panel consisted of 16 x 16 pixels and each pixel consisted of 2 OTFTs, 1 Capacitor and 1 OLED. The pixels successfully worked in terms of current magnitude supplied to OLED and the control ability of driving and switching OTFTs.

• 한국정보디스플레이학회:학술대회논문집
• /
• 2007.08b
• /
• pp.1685-1687
• /
• 2007

Excellent moving picture performance of plasma TV has been confirmed through the study and measurement using the method proposed by the APDC (Advanced PDP Development Center Corporation). Full-HD Plasma TVs with pixels of $1920{\times}1080[1]-[2]$, showed more than 900 TV-lines of resolution out of maximum scale of 1080, while typical LCD of the same pixel count showed only 1/3 of the plasma's performance. Moreover, even the latest 120Hz models of $1920{\times}1080$ LCD do not much PDPs with pixels of $1024{\times}768$, or $1366{\times}786$ in moving picture resolution. The measurement system proposed by the APDC showed very good agreement with subjective tests.

• Proceedings of the KSRS Conference
• /
• 2006.03a
• /
• pp.7-14
• /
• 2006

Detection and identification of targets from remotely sensed imagery are of great interest for civilian and military application. This paper presents an algorithm for target detection in high spatial resolution imagery based on the spectral and the dimensional characteristics of the reference target. In this algorithm, the spectral and the dimensional information of the reference target is extracted automatically from the sample image of the reference target. Then in the entire image, the candidate target pixels are extracted based on the spectral characteristics of the reference target. Finally, groups of candidate pixels which form isolated spatial objects of similar size to that of the reference target are extracted as detected targets. The experimental test results showed that even though the algorithm detected spatial objects which has different shape as targets if the spectral and the dimensional characteristics are similar to that of the reference target, it could detect 97.5% of the targets in the image. Using hyperspectral image and utilizing the shape information are expected to increase the performance of the proposed algorithm.