• Journal of the Korea Society of Computer and Information
• /
• v.22 no.7
• /
• pp.47-54
• /
• 2017

Recently, correlation filter based trackers have shown excellent tracking performance and computational efficiency. In order to enhance tracking performance in the correlation filter based tracker, search area which is image patch for finding target must include target. In this paper, two methods to discriminatively represent target in the search area are proposed. Firstly, search area location is estimated using pyramidal Lucas-Kanade algorithm. By estimating search area location before filtering, fast motion target can be included in the search area. Secondly, we investigate multi-channel Local Edge Pattern(LEP) which is insensitive to illumination and noise variation. Qualitative and quantitative experiments are performed with eight dataset, which includes ground truth. In comparison with method without search area estimation, our approach retain tracking for the fast motion target. Additionally, the proposed multi-channel LEP improves discriminative performance compare to existing features.

• Journal of Institute of Control, Robotics and Systems
• /
• v.18 no.7
• /
• pp.673-679
• /
• 2012

This paper introduces robust tracking algorithm for fast and erratic moving object. CAMSHIFT algorithm has less computation and efficient performance for object tracking. However, the method fails to track a object if it moves out of search window by fast velocity and/or large movement. The size of the search window in CAMSHIFT algorithm should be selected manually also. To solve these problems, we propose an efficient prediction technique for fast movement of object using Kalman Filter with automatic initial setting and variable configuration technique for search window. The proposed method is compared to the traditional CAMSHIFT algorithm for searching and tracking performance of objects on test image frames.

• Journal of the Korea Convergence Society
• /
• v.8 no.9
• /
• pp.249-254
• /
• 2017

The purpose of this study is to provide a way to develop the information quality of mobile real estate service and improve the false property through search filter service. I compared the cases of typical mobile real estate apps that are currently in use and analyzed the usage and service status by adopting 13 search filters that are the first priority when searching for real estate information. As a result, it has been shown that the provision of high quality search filter service is required to improve the information quality in order to provide users with quick and reliable information organically at any time and anywhere according to the definition of mobile real estate service. As a result, it is more prejudiced than the countermeasures against false fraudulent products, and it is satisfied with the needs of the users and provides the improved search filter service, so that it can be trusted with the definition of mobile real estate brokerage service, Providing quality services can be said to prevent the victims of fraudulent products.

• Journal of KIISE
• /
• v.42 no.8
• /
• pp.961-971
• /
• 2015

As a representative area-decomposed algorithm, an area-based quad-trie (AQT) has an issue of search performance. The search procedure must continue to follow the path to its end, due to the possibility of the higher priority-matching rule, even though a matching rule is encountered in a node. A leaf-pushing AQT improves the search performance of the AQT by making a single rule node exist in each search path. This paper proposes a new algorithm to further improve the search performance of the leaf-pushing AQT. The proposed algorithm implements a leaf-pushing AQT using a hash table and an on-chip Bloom filter. In the proposed algorithm, by sequentially querying the Bloom filter, the level of the rule node in the leaf-pushing AQT is identified first. After this procedure, the rule database, which is usually stored in an off-chip memory, is accessed. Simulation results show that packet classification can be performed through a single hash table access using a reasonable sized Bloom filter. The proposed algorithm is compared with existing algorithms in terms of the memory requirement and the search performance.

• Journal of the Korea Society of Computer and Information
• /
• v.22 no.12
• /
• pp.117-123
• /
• 2017

The personalized search algorithm is a search system that analyzes the user's IP, cookies, log data, and search history to recommend the desired information. As a result, users are isolated in the information frame recommended by the algorithm. This is called 'Filter bubble' phenomenon. Most of the personalized data can be deleted or changed by the user, but data stored in the service provider's server is difficult to access. This study suggests a way to neutralize personalization by keeping on sending random query words. This is to confuse the data accumulated in the server while performing search activities with words that are not related to the user. We have analyzed the rank change of the URL while conducting the search activity with 500 random query words once using the personalized account as the experimental group. To prove the effect, we set up a new account and set it as a control. We then searched the same set of queries with these two accounts, stored the URL data, and scored the rank variation. The URLs ranked on the upper page are weighted more than the lower-ranked URLs. At the beginning of the experiment, the difference between the scores of the two accounts was insignificant. As experiments continue, the number of random query words accumulated in the server increases and results show meaningful difference.

• Journal of Korea Multimedia Society
• /
• v.21 no.3
• /
• pp.369-381
• /
• 2018

Nowadays a personalization algorithm is gaining huge attention. It gives users selective information which is helpful and interesting in a deluge of information based on their past behavior on the internet. However there is also a fatal side effect that the user can only get restricted information on restricted topics selected by the algorithm. Basically, the personalization algorithm makes users have a narrower perspective and even stronger bias because users have less chances to get views of opponent. Eli Pariser called this problem the 'filter bubble' in his book. It is important to understand exactly what a filter bubble is to solve the problem. Therefore, this paper shows how much Google's personalized search algorithm influences search result through an experiment with deep neural networks acting like users. At the beginning of the experiment, two Google accounts are newly created, not to be influenced by the Google's personalized search algorithm. Then the two pure accounts get politically biased by two methods. We periodically calculate the numerical score depending on the character of links and it shows how biased the account is. In conclusion, this paper shows the formation process of filter bubble by a personalization algorithm through the experiment.

• Journal of Advanced Navigation Technology
• /
• v.16 no.3
• /
• pp.545-552
• /
• 2012

In this paper, we propose Modified Multi CAMShift Algorithm based on Kalman filter and Weighted Search Windows(KWMCAMShift) that extracts skin color area and tracks several human body parts for real-time human tracking system. We propose modified CAMShift algorithm that generates background model, extracts skin area of hands and head, and tracks the body parts. Kalman filter stabilizes tracking search window of skin area due to changing skin area in consecutive frames. Each occlusion areas is avoided by using weighted window of non-search areas and main-search area. And shadows are eliminated from background model and intensity of shadow. The proposed KWMCAMShift algorithm can estimate human pose in real-time and achieves 96.82% accuracy even in the case of occlusions.

• Journal of KIISE:Databases
• /
• v.36 no.2
• /
• pp.84-98
• /
• 2009

Various algorithms and architectures for IP address lookup have been studied to improve forwarding performance in the Internet routers. Previous IP address lookup architecture using Bloom filter requires a separate Bloom filter as well as a separate hash table in each prefix length, and hence it is not efficient in implementation complexity. To reduce the number of hash tables, it applies controlled prefix expansion, but prefix duplication is inevitable in the controlled prefix expansion. Previous parallel multiple-hashing architecture shows very good search performance since it performs parallel search on tables constructed in each prefix length. However, it also has high implementation complexity because of the parallel search structure. In this paper, we propose a new IP address lookup architecture using all-length Bloom filter and all-length multiple hash table, in which various length prefixes are accomodated in a single Bloom filter and a single multiple hash table. Hence the proposed architecture is very good in terms of implementation complexity as well as search performance. Simulation results using actual backbone routing tables which have $15000{\sim}220000$ prefixes show that the proposed architecture requires 1.04-1.17 memory accesses in average for an IP address lookup.

• 제어로봇시스템학회:학술대회논문집
• /
• 2004.08a
• /
• pp.1334-1337
• /
• 2004

This paper presents a novel method to speed up neural network (NN) based face detection systems. NN-based face detection can be viewed as a classification and search problem. The proposed method formulates the search problem as an integer nonlinear optimization problem (INLP) and expands the basic particle swarm optimization (PSO) to solve it. PSO works with a population of particles, each representing a subwindow in an input image. The subwindows are evaluated by how well they match a NN-based face filter. A face is indicated when the filter response of the best particle is above a given threshold. To achieve better performance, the influence of PSO parameter settings on the search performance was investigated. Experiments show that with fine-adjusted parameters, the proposed method leads to a speedup of 94 on 320${\times}$240 images compared to the traditional exhaustive search method.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.21 no.3
• /
• pp.684-691
• /
• 1996

Since the period of a spreading code in DS/SS communication systems is generally long, it is necessary to make the code acquisition as fast as possible. The code acquisition time can be sued as a measure to evaluate the performance of code acquisition systems. The search rate of serial search codee acquisition system used in the coventional CDMA cellular system is lower than that of the matched filter technique. In order to reduce the code acquisition time, this paper proposes hybrid code acquisition system composed filters combined with serial search blocks. In the proposed system, the matched filter sweeps possible code phases fast and the acquired phase information is verified by the serial search block. The mean and the variance of its acquisition time are calculated and compared with those of double dwell serial search system. The results indicate better performance of the proposed system by yielding its small vaues of the mean and the variance of code acquisition time.