• IEMEK Journal of Embedded Systems and Applications
• /
• v.16 no.5
• /
• pp.179-186
• /
• 2021

For stereo matching based on deep learning, the design of network structure is crucial to the calculation of matching cost, and the time-consuming problem of convolutional neural network in image processing also needs to be solved urgently. In this paper, a method of stereo matching using sparse loss volume in parallax dimension is proposed. A sparse 3D loss volume is constructed by using a wide step length translation of the right view feature map, which reduces the video memory and computing resources required by the 3D convolution module by several times. In order to improve the accuracy of the algorithm, the nonlinear up-sampling of the matching loss in the parallax dimension is carried out by using the method of multi-category output, and the training model is combined with two kinds of loss functions. Compared with the benchmark algorithm, the proposed algorithm not only improves the accuracy but also shortens the running time by about 30%.

• The KIPS Transactions:PartA
• /
• v.9A no.3
• /
• pp.295-300
• /
• 2002

In this paper, we apply the 3-dimension algorithm to the random interleaver and s-random interleaver and analyze the performance of the turbo code system with random interleaver (or s-random interleaver). In general, the performance of interleaver is determined by minimum distance between neighbor data, thus we could improve the performance of interleaver by increasing the distance of the nearest data. The interleaver using 3-dimension algorithm has longer minimum distance and average distance compared to existing random-interleaver (s-random interleaver) because the output data is generated randomly from 3-dimension storage. To verify and compare the performance of our proposed system, the computer simulations have been performed in turbo code system under gaussian noise environment.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.23 no.57
• /
• pp.51-62
• /
• 2000

In this study, the psycho-physiological response of drivers was investigated in terms of EEG(Electroencephalogram), especially with the fractal dimensions computed by Chaotic algorithm. The Chaotic algorithm Is well Known to sensitively analyze the non-linear information such as brain waves. An automobile with a fully equipped data acquisition system was used to collect the data. Ten healthy subjects participated in the experiment. EEG data were collected while subjects were driving the car between Won-ju and Shin-gal J.C. on Young-Dong highway The results were presented in terms of 3-Dimensional attractor to confirm the chaotic nature of the EEG data. The correlation dimension and fractal dimension were calculated to evaluate the complexity of the brain activity as the driving duration changes. In particular, the fractal dimension indicated a difference between the driving condition and non-driving condition while other spectral variables showed inconsistent results. Based upon the fractal dimension, drivers processed the most information at the beginning of the highway driving and the amount of brain activity gradually decreased and stabilized. No particular decrease of brain activity was observed even after 100 km driving. Considering the sensitivity and consistency of the analysis by Chaotic algorithm, the fractal dimension can be a useful parameter to evaluate the psycho-physiological responses of human brain at various driving conditions.

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 2003.04a
• /
• pp.255-259
• /
• 2003

Moire topography method isa well-known non-contacting 3-D measurement method as afast non-contact test for three-dimension shape measuring method. Recently, it's important to study the automatic three-dimension measurement by moire topography because it is frequently applied to the reverse engineering , the medical , the entertainment fields. Three-dimension measurement using projection of moire topography is very attractive because of its high measuring speed and high sensitivity. In this paper, the classical moire method is computerized-so called digital moire when a virtual grating pattern is projected on a surface, the captured image by the CCD camera has three-dimension information of the objects. The moire image can be obtained through a simple image processing and a reference grating pattern. and it provides similar results without physical grating pattern. digital projection moire topography turn out to be very effective for the three-dimension measurement of objects. Using different N-bucket algorithm method of digital projection moire topography is tested to measuring object with the 2-ambiguity problem. Experimental results prove that the proposed scheme is capable of finding measurement errors that decreased more by using the four-three step algorithm method instead of the same step in the phase shifting of different pitch.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1992.04a
• /
• pp.153-156
• /
• 1992

디지탈 영상처리는 60년대 초부터 사용되어 왔으며 많은 양의 DATA를 처리해야 한다는 점과 일반적으로 적용될 수 있도록 충분히 빠르고 안정된 Algorithm이 요구된다는 점에서 상당히 느리게 발전되어 왔으며, 70년대에 들어서 Computer Vision이라는 학문이 상당한 연구과제로 부각되어 2 Dimension의 해석 및 인식을 위한 Algorithm들이 각종 자료를 통하여 발표되었고 80년대에 들어서부터 3 Dimension의 인식을 위한 Algorithme들이 연구과제로 진행되고 있다. 본 프로잭트는 그동안 국내에서 어려웠던 브라운관의 Crack 및 Size 감지의 경우 99.8%의 자동감지 효과를 구현하였다.

• Journal of Korean Port Research
• /
• v.14 no.3
• /
• pp.303-312
• /
• 2000

In this paper, we describe an image processing algorithm which is able to recognize the road lane. This algorithm performs to recognize the interrelation between AGV and the other vehicle. We experimented on AGV driving test with color CCD camera which is setup on the top of vehicle and acquires the digital signal. This paper is composed of two parts. One is image preprocessing part to measure the condition of the condition of the lane and vehicle. This finds the information of lines using RGB ratio cutting algorithm, the edge detection and Hough transform. The other obtains the situation of other vehicles using the image processing and viewport. At first, 2 dimension image information derived from vision sensor is interpreted to the 3 dimension information by the angle and position of the CCD camera. Through these processes, if vehicle knows the driving conditions which are lane angle, distance error and real position of other vehicles, we should calculate the reference steering angle.

• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.339-339
• /
• 2000

In this parer, a method of measuring the size of defects on the wall and restructing the defect image of 3-dimension is developed based on the tracking algorithm of a camera position which uses the inclination angle of line slit beam for overcoming the difficulty of the corresponding problem identifying the image point in the both image. In the experiments, an algorithm for estimating the horizontal angle of CCD camera is presented and validated by applying it to the measurement of area and length under the variations of both the distance and the angle of CCD camera. And its performance is compared to that of the rotating and mapping method of image which has the Euclidian distance.

• Journal of the Korean Institute of Navigation
• /
• v.16 no.3
• /
• pp.77-85
• /
• 1992

Camera arrangement for depth and image correspondence is very important to the computer vision. Two conventional comera arrangements for stereo computer vision are lateral model and axial motion model. In this paper, using the axial motion stereo camera model, the algorithm for camera focal length measurement and the surface smoothness with the radiance-irradiance is proposed fro 3-dimensional image correspondence on stereo computer vision. By adapting the above algorithm, camera focal length can be measured precisely and the resolution of 3-dimensional image correspondence has been improved comparing to that of the axial motion model without the radiance-irradiance relation.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2006.05a
• /
• pp.957-960
• /
• 2006

In this paper, we studied the location estimation algorithm of a spatial 3 dimension which extend the location estimation algorithm of a plane 2 dimension in 2.45GHz band RTLS(Real time location system). We used TDOA scheme which need not a time of transmission information of the tag and estimated 3 dimension coordinates. Also, estimated intersection of hyperbolic curve to X, Y coordinate of the tag at 2D coordinates searching area, $300m\times300m$ and LOS propagation environments. And, we estimated Z coordinate ultimately using X, Y coordinate. The location estimation algorithm of a spatial 3 dimension satisfies the RTLS specification requirement, 3m radius accuracy. From the result, we confirm that the location of tag which similar to actual coordinate in the case to an ideal received offset. However, we verified that the location of tag which escapes from a radius 3m within error range when received offset increased. Therefore, as the future work we are consider enhanced location accuracy of a spatial 3 dimension in RTLS system which using the decrease scheme of reader offset or the discriminate scheme of the estimation location.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.21 no.3
• /
• pp.19-25
• /
• 1984

This paper suggests a new algorithm which minimizes the bit dimension of control memory considering the flexibility for reprogramming in the microprogrammed digital systems. The algorithm can not only minimize the bit dimension but also compensate the loss of flexibility by giving a highest MOP the unique field in the process of bit minimization. Also, programming the algorithm, that result and previous works are compared and reviewed. Then the bit reduction and the improvement of flexibility are proved.