• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.11 no.5
• /
• pp.1677-1684
• /
• 2010

Recently, due to the development of mobile phone and supply of smart phone, many contents have been developed. Especially, since the small-sized cameras are equiped in mobile devices, people are interested in the image based contents development, and it also becomes important part in their practical use. Among them, the character recognition system can be widely used in the applications such as blind people guidance systems, automatic robot navigation systems, automatic video retrieval and indexing systems, automatic text translation systems. Therefore, this paper proposes a system that is able to extract text area from the natural images captured by smart phone camera. The individual characters are recognized and result is output in voice. Text areas are extracted using Adaboost algorithm and individual characters are recognized using error back propagated neural network.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• v.9 no.2
• /
• pp.195-199
• /
• 2005

Semi automatic welding method to use carriage for welding at large size block manufacture welding process of present shipbuilding industry is used much. Carriage is device that transfer welding torch in horizontal fillet weld here, but because it is no function that chase welding like robot welding method, use can be impossible in curved line welding, and simply use in straight line welding. Also, because it is no function that chase welding, though welding mistake corrects this happening often in straight line welding, much times and expense are cost. Added welding chase sensor and 80C196KC microcontroller that use strain gauge to carriage that is using present in paper that see hereupon and manufacture a private line model and container box model welding because developing system that can chase welding automatically straight line and curved line welding establishing and investigate about chase phenomenon.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.17 no.4
• /
• pp.106-115
• /
• 2003

In crossing stereo camera system as a security equipment system, a new algorithm for the tracking method of the moving object and the measurement method of distance of the object from cameras to moving object is proposed. The stereo input images are matched by a pixel for the moving object extraction, then the moving elements obtained from an adaptive threshold's value for an error correction and the moving object area extracted by noise eliminative algorithm. In addition to, the moving object traced by control pan/tilt of cameras using a coordinate value of the moving object, and the measurement method of distance to moving object by using special features of the crossing stereo camera system is proposed. From the experimental results, the proposed algorithm was found to be able to measure of the distance and trace for moving object in a stereo security camera system. The error rate of a distance measurement be existed within 6%. Moreover in implementation the proposed algorithm, the stereo security system, the automatic moving robot system and the stereo remote control system will be applied practical application.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.9
• /
• pp.1032-1037
• /
• 2019

In the existing pharmacy automation system, quantity control was carried out only in the memory without first measuring the quantity of the medicine and during the operation of the facility. As a result, the number of drugs that have already been deducted during operation at the time of a facility error has not been managed. Serious problems are caused by important medicines that need to be managed because of lack of quantity control. In addition, the user had to refill the drug when it was exhausted because the user could not know the amount of the drug outside the facility. When the drug was out of stock, the facility was operating with loss of time to restart and the time to replenish the drug. Thus, in this paper, we designed and applied system to control the quantity of medicines frequently by adding displacement sensors to the robot of the facility.

• Journal of the Institute of Convergence Signal Processing
• /
• v.21 no.3
• /
• pp.113-120
• /
• 2020

With the recent development of communication equipment, the demand for communication equipment is gradually increasing. Accordingly, various signal processing has been studied. In the case of an image, noise removal is an indispensable step because noise propagation problems may occur if noise is not removed in the pre-processing process. Salt and Pepper noise is a typical impulse noise with two extremes. Various studies have been conducted to remove such noise, and there are CWMF, MF and MMF. However, the existing methods are somewhat insufficient in the high-density noise region. Therefore, in this study, we have proposed an algorithm that filters the size of the mask according to the number of noises inside the 7×7 mask and filters it with a modified switching filter using the histogram distribution of the image. In the case of the proposed algorithm, noise can be effectively removed in a high-density noise region. For objective judgment, PSNR was used to compare and analyze with existing algorithms.

• Journal of Advanced Navigation Technology
• /
• v.27 no.2
• /
• pp.182-189
• /
• 2023

Depth estimation is a key technology in 3D map generation for autonomous driving of vehicles, robots, and drones. The existing sensor-based method has high accuracy but is expensive and has low resolution, while the camera-based method is more affordable with higher resolution. In this study, we propose self-attention-based unsupervised monocular depth estimation for UAV camera system. Self-Attention operation is applied to the network to improve the global feature extraction performance. In addition, we reduce the weight size of the self-attention operation for a low computational amount. The estimated depth and camera pose are transformed into point cloud. The point cloud is mapped into 3D map using the occupancy grid of Octree structure. The proposed network is evaluated using synthesized images and depth sequences from the Mid-Air dataset. Our network demonstrates a 7.69% reduction in error compared to prior studies.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.4
• /
• pp.144-159
• /
• 2014

Techniques for making a single panoramic image using multiple pictures are widely studied in many areas such as computer vision, computer graphics, etc. The panorama image can be applied to various fields like virtual reality, robot vision areas which require wide-angled shots as an useful way to overcome the limitations such as picture-angle, resolutions, and internal informations of an image taken from a single camera. It is so much meaningful in a point that a panoramic image usually provides better immersion feeling than a plain image. Although there are many ways to build a panoramic image, most of them are using the way of extracting feature points and matching points of each images for making a single panoramic image. In addition, those methods use the RANSAC(RANdom SAmple Consensus) algorithm with matching points and the Homography matrix to transform the image. The SURF(Speeded Up Robust Features) algorithm which is used in this paper to extract featuring points uses an image's black and white informations and local spatial informations. The SURF is widely being used since it is very much robust at detecting image's size, view-point changes, and additionally, faster than the SIFT(Scale Invariant Features Transform) algorithm. The SURF has a shortcoming of making an error which results in decreasing the RANSAC algorithm's performance speed when extracting image's feature points. As a result, this may increase the CPU usage occupation rate. The error of detecting matching points may role as a critical reason for disqualifying panoramic image's accuracy and lucidity. In this paper, in order to minimize errors of extracting matching points, we used $3{\times}3$ region's RGB pixel values around the matching points' coordinates to perform intermediate filtering process for removing wrong matching points. We have also presented analysis and evaluation results relating to enhanced working speed for producing a panorama image, CPU usage rate, extracted matching points' decreasing rate and accuracy.

• Cartoon and Animation Studies
• /
• s.20
• /
• pp.47-61
• /
• 2010

The Uncanny Valley is a hypothesis introduced by a roboticist, Masahiro Mori. The theory holds that as robots and other facsimiles of humans look and act more human-like, the emotional response from a human spectator to them will become increasingly empathic, until they approach realistic similarity to humans but not quite exactly like them, when those images stop being likable and instead become repulsive and "uncanny." Although the scientific validity of this hypothesis is still debatable, it is true that many spectators has tendency to reject hyper-realistic animation. For that reason, the great success of the recent movie Avatar is remarkable. Computer animations that try to represent humans in hyper-realistic way fall into a paradox: successful result images would look too mundane and unsuccessful ones look "wrong". On top of that, audience would expect from hyper-realistic animation to see same level of acting of live-action films, which is usually not the case. The subtle differences between digital characters and real actors irritate spectators and hinder them in absorbtion. The biggest difference between Avatar and other hyper-realistic animation is the referent of their representation. While Avatar pursues a similar hyper-realistic visual style, the referents of the representation shown in this movie are totally alien to us thus impossible to compare with the real counterparts. Unlike characters in other hyper-realistic animations who had to fight an uphill battle with great actors in real life, those in Avatar have criteria of their own. That is why this movie can be free from the dreadful valley and let the audience absorb into the spectacle.