• Title/Summary/Keyword: 유사 중복 검출

Search Result 22, Processing Time 0.025 seconds

A Fast and Efficient Haar-Like Feature Selection Algorithm for Object Detection (객체검출을 위한 빠르고 효율적인 Haar-Like 피쳐 선택 알고리즘)

  • Chung, Byung Woo;Park, Ki-Yeong;Hwang, Sun-Young
    • The Journal of Korean Institute of Communications and Information Sciences
    • /
    • v.38A no.6
    • /
    • pp.486-491
    • /
    • 2013
  • This paper proposes a fast and efficient Haar-like feature selection algorithm for training classifier used in object detection. Many features selected by Haar-like feature selection algorithm and existing AdaBoost algorithm are either similar in shape or overlapping due to considering only feature's error rate. The proposed algorithm calculates similarity of features by their shape and distance between features. Fast and efficient feature selection is made possible by removing selected features and features with high similarity from feature set. FERET face database is used to compare performance of classifiers trained by previous algorithm and proposed algorithm. Experimental results show improved performance comparing classifier trained by proposed method to classifier trained by previous method. When classifier is trained to show same performance, proposed method shows 20% reduction of features used in classification.

Thermal Imagery-based Object Detection Algorithm for Low-Light Level Nighttime Surveillance System (저조도 야간 감시 시스템을 위한 열영상 기반 객체 검출 알고리즘)

  • Chang, Jeong-Uk;Lin, Chi-Ho
    • The Journal of The Korea Institute of Intelligent Transport Systems
    • /
    • v.19 no.3
    • /
    • pp.129-136
    • /
    • 2020
  • In this paper, we propose a thermal imagery-based object detection algorithm for low-light level nighttime surveillance system. Many features selected by Haar-like feature selection algorithm and existing Adaboost algorithm are often vulnerable to noise and problems with similar or overlapping feature set for learning samples. It also removes noise from the feature set from the surveillance image of the low-light night environment, and implements it using the lightweight extended Haar feature and adaboost learning algorithm to enable fast and efficient real-time feature selection. Experiments use extended Haar feature points to recognize non-predictive objects with motion in nighttime low-light environments. The Adaboost learning algorithm with video frame 800*600 thermal image as input is implemented with CUDA 9.0 platform for simulation. As a result, the results of object detection confirmed that the success rate was about 90% or more, and the processing speed was about 30% faster than the computational results obtained through histogram equalization operations in general images.

Microbial Quality of Fresh Vegetables and Fruits in Seoul, Korea (국내 신선 채소류의 미생물 오염 특성)

  • Hong, Chae-Kyu;Seo, Young-Ho;Choi, Chae-Man;Hwang, In-Suk;Kim, Moo-Sang
    • Journal of Food Hygiene and Safety
    • /
    • v.27 no.1
    • /
    • pp.24-29
    • /
    • 2012
  • A total of 187 samples of leafy vegetables and fruits were acquired at traditional markets and department stores in Seoul, Korea. Samples were tested for microorganism distributions and for the presence of pathogenic bacteria. The aerobic mesophilic counts ranged between 2.5 and 9.4 log CFU/g, with the highest count recorded from the dropwort. Counts of psychrotrophic microorganisms were as high as those of the mesophilic microorganisms. Total coliform populations between 1.0 and 7.8 log CFU/g were found in 90.9% of the samples. Microbiological counts for fruits were very low. $Escherichia$ $coli$ was isolated in 24 (12.8%) samples. $Staphylococcus$ $aureus$ and $Clostridium$ $perfringens$ contamination were found in 15 (8.0%) and 20 (10.7%) samples. $Salmonella$ species and $Listeria$ $monocytogenes$ were detected in 2.7 and 0.5% of samples, respectively. Among the total 187 samples, 8 samples were contaminated by more than two pathogens. $E.$ $coli$ O157:H7 was not detected in any of the samples. The microbial contamination levels determined in the present study may be used as the primary data to execute microbial risk assessment of fresh vegetables and fruits.

DAL-based Hierarchical Slicing Thinning Algorithm (DAL을 고려한 계층적 슬라이싱 세선화 알고리즘)

  • Choi, Won-Seok;Koh, Chang-Seok;Jang, Kyung-Sik
    • Proceedings of the Korean Institute of Information and Commucation Sciences Conference
    • /
    • 2009.10a
    • /
    • pp.941-944
    • /
    • 2009
  • In this paper, we present a improved algorithm for thinning vein pattern. The vein image scanned using near infra-red is sliced in the horizontality or verticality direction, and then average value is gained from each slice image. Using curvature value changed from average value, search candidate part and extract vein pattern through re-searched candidate part in overlapping area. For the extracted vein pattern connecting, we proposed a DAL method that verify distance, angle and luminosity on pattern before tracking pattern extracted. The proposed improved thinning method is helpful for accurate connecting of vein pattern more than method used distance material in candidate part.

  • PDF

Prevalence of Toxin Genes and Antibiotic Resistance Profiles of Vibrio vulnificus strains isolated from Jeju Island (제주도에서 분리된 비브리오패혈증균의 독소 유전자 분포 및 항생제 내성)

  • Eunok Kang;Man Jae Cho;Ye-Seul Heo;Eun A Koh
    • Journal of Food Hygiene and Safety
    • /
    • v.38 no.5
    • /
    • pp.381-389
    • /
    • 2023
  • Vibrio vulnificus, the most fatal waterborne and foodborne pathogens of 50% fatality rate in the world, is common in seawater and occurs particularly in warmer months. In this study, we investigated the toxin genes using reverse transcription-polymerase chain reaction (RT-PCR), antibiotic resistance status using Vitek, and genetic characteristics using pulsed-field gel electrophoresis (PFGE) of different V. vulnificus strains isolated from the Jeju Island seawater, distribution fishery products, and fish tanks. We examined a total of 487 samples and isolated a total of 46 strains (including overlapping strains) of V. vulnificus, 44 strains from seawater and 1 strain each from fishery products and fish tank. We detected toxin gene vvhA in all 46 strains and rtxA, viu in 8 strains (17.4%) and 9 strains (19.6%) strains, respectively. Antibiotic resistance tests indicated 100% resistance to cefoxitin antibiotics. The PFGE analysis of the 46 strains identified a total of 6 types showed 100% homology and the degree of similarity was 81.3-98.0%; however, there were no similarity between the regions and samples. These results indicate that V. vulnificus isolated from the seawater, fishery products, and fish tanks should be continuously monitored as cases of food poisoning caused by V. vulnificus with toxin genes have been reported in Jeju Island.

Weighted Disassemble-based Correction Method to Improve Recognition Rates of Korean Text in Signboard Images (간판영상에서 한글 인식 성능향상을 위한 가중치 기반 음소 단위 분할 교정)

  • Lee, Myung-Hun;Yang, Hyung-Jeong;Kim, Soo-Hyung;Lee, Guee-Sang;Kim, Sun-Hee
    • The Journal of the Korea Contents Association
    • /
    • v.12 no.2
    • /
    • pp.105-115
    • /
    • 2012
  • In this paper, we propose a correction method using phoneme unit segmentation to solve misrecognition of Korean Texts in signboard images using weighted Disassemble Levenshtein Distance. The proposed method calculates distances of recognized texts which are segmented into phoneme units and detects the best matched texts from signboard text database. For verifying the efficiency of the proposed method, a database dictionary is built using 1.3 million words of nationwide signboard through removing duplicated words. We compared the proposed method to Levenshtein Distance and Disassemble Levenshtein Distance which are common representative text string comparison algorithms. As a result, the proposed method based on weighted Disassemble Levenshtein Distance represents an improvement in recognition rates 29.85% and 6% on average compared to that of conventional methods, respectively.

Effectiveness of Microsatellite Markers for Parentage Analysis of Giant Grouper (Epinephelus lanceolatus) Broodstock (Microsatellite 마커를 이용한 대왕바리(Epinephelus lanceolatus) 친어 집단의 가계도 분석 효율)

  • Kim, Keun-Sik;Noh, Choong Hwan;Sade, Ahemad;Bang, In-Chul
    • Korean Journal of Ichthyology
    • /
    • v.27 no.1
    • /
    • pp.10-15
    • /
    • 2015
  • Giant grouper (Epinephelus lanceolatus) is a endangered species considered as a vulnerable grade-organism in the International Union for Conservation of Nature (IUCN) red list. As a fundamental baseline study for establishing a giant grouper broodstock management system, the efficiency for parentage analysis was evaluated by using microsatellite makers previously available in this species. The eight microsatellites generated a total 52 alleles from 32 individuals, the mean expected heterozygosity was 0.663, and mean inbreeding coefficient was 0.011, consequently suggesting that the present broodstock has retained the high level of genetic diversity. However, our analysis also recommended the collection of more broodfish for more stable brood line, since the estimated value of the effective population size was proven to be 35. The average probability of identity was $6.85{\times}10^{-11}$. NE-2P and NE-PP of paternity non-exclusion probabilities were 0.00835 and 0.00027, respectively. As the result of principle coordinate analysis, the genotype of broodstock was not overlapped, suggesting that the management system of giant grouper based on eight selected microsatellite markers might be effective, although further validation with extended number of broodfish might also be needed in future. Data of present study could be a useful basis to avoid the unwanted selection of broodfish that possess close genetic relationship with current broodstock, and consequently to establish effective broodstock management system allowing the production of progeny with high genetic diversity.

A Missense Mutation in Exon 5 of the Bovine Growth Hormone Gene (소 성장호르몬 유전자의 Exon 5번에서의 새로운 다형성 연구)

  • Yoon, D. H.;Kim, T. H.;Lee, K. H.;Park, E. W.;Lee, H. K.;Cheong, I. C.;Hong, K. C.
    • Journal of Animal Science and Technology
    • /
    • v.45 no.1
    • /
    • pp.13-22
    • /
    • 2003
  • Growth Hormone (GH) gene is a member of gene family through the evolutionary process from a small common ancestral gene by a series of gene duplications. The role of the GH in growth and performance controls has been extensively studied in human, mice and livestock. Many researchers have considered GH as a strong candidate gene for evaluation of genetic polymorphisms that could be associated with economic traits in cattle. We report here a novel missense mutation within the exon 5 of the bovine Growth Hormone (bGH) gene. We could amplified 522 bp fragments from eight unrelated Hanwoo cattle by PCR, then, subsequently cloned and sequenced. An Msp I RFLP corresponding to a C to T transition was observed at position 2258 nt. From this result, we could predict a missense mutation (Arg to Trp) at codon 166 in a highly conserved region among many mammals. Codominant Mendelian segregation of the two alleles, Msp I (+) and Msp I (-), was observed in two full-sib F2 families (n = 32, African taurine Bos taurus ${\times}$ African zebu Bos indicus) and eight half-sib Hanwoo families. For the availability of genetic marker, we have performed PCR-RFLP with a large number of individual animals from 15 different cattle breeds (European and Asian taurines, and African indicines). Consideration of breed frequencies of Msp I (-) allele in relation to breed type and their geographic origins, shows higher frequencies in humped breeds or Asian cattle breeds than in humpless or European breeds. This result indicates that the missense mutation can be contributed the functional significance such as the signal transduction through the receptor binding, also may be used as a marker for selection of the economic traits in Hanwoo.

Occurrence of Fungi and Fusarium Mycotoxins in the Rice Samples from Rice Processing Complexes (미곡종합처리장 곡류시료 내 곰팡이 오염과 Fusarium 독소 발생)

  • Lee, Theresa;Lee, Soohyung;Kim, Lee-Han;Ryu, Jae-Gee
    • Research in Plant Disease
    • /
    • v.20 no.4
    • /
    • pp.289-294
    • /
    • 2014
  • Rice samples including paddy, husk, brown rice, blue-tinged rice, broken rice, discolored rice and polished rice were collected from rice processing complexes(RPC) nationwide to determine the contamination of fungi and mycotoxins on rice during 2010-2013. Among the samples, paddy rice had the highest frequencies of fungal and Fusarium occurrence, and the frequencies decreased along with milling as husk was the next. Blue-tinged rice or discolored rice was similar with brown rice for fungal occurrence, and polished rice showed the lowest frequency. Among Fusarium species, F. graminearum species complex occupied 87% in 2012 but did 35-39% in 2011 and 2013. Aspergillus and Penicillium species appeared at low frequency in most samples but occurred at higher frequency in certain RPC samples. Alternaria, Nigrospora, and Epicoccum species occurred similarly to the pattern of total fungi. The rice samples from 2010-2012 were analyzed for the occurrence of Fusarium mycotoxins including deoxynivalenol, nivalenol, zearalenone, and other trichothecenes. The most highly contaminated sample was discolored rice in terms of frequency, level, ratio of simultaneous contamination with multiple toxins, followed by blue-tinged rice.

Physicochemical Properties and Microbial Analysis of Korean Solar Salt and Flower of Salt (한국산 꽃소금과 천일염의 이화학적 특성 및 미생물 분석)

  • Lee, Hye Mi;Lee, Woo Kyoung;Jin, Jung Hyun;Kim, In Cheol
    • Journal of the Korean Society of Food Science and Nutrition
    • /
    • v.42 no.7
    • /
    • pp.1115-1124
    • /
    • 2013
  • The present study was conducted to ensure the diversity of domestic solar salt by analyzing the composition and microbiological characteristics of solar salt (from Docho island: DS) and the flower of salt produced in different Korean salt flats (Sinui island: SF, Bigum island: BF, and Docho island: DF). The analyses showed that the moisture content of the three types of flower of salt and solar salt ranged from 10.54~13.82% and NaCl content ranged from 78.81~84.61%. The mineral content of those salts ranged from 3.57~5.51%. The content of insoluble matter in these salts was $0.01{\pm}0.00{\sim}0.05{\pm}0.00%$. The sand content of these salts was $0.01{\pm}0.01{\sim}0.03{\pm}0.01%$. By Hunter's color value analysis, the color of the flower of salt was brighter and whiter than solar salt. The salinity of the flower of salt was a little higher than solar salt as well. The magnesium and potassium ion content of DF was $9,886.72{\pm}104.78mg/kg$ and $2,975.23{\pm}79.73mg/kg$, respectively, which was lower than the content in SF, BF, and DS. The heavy metal content of all salts was acceptable under the Korean Food Sanitation Law. The flower of salt was confirmed to be sweeter and preferable to solar salt. More than 80% of the solar salt crystals were 2~3 mm in size, whereas crystals from the flower of salt were 0.5~2 mm in size. The bacterial diversity of DF and DS were investigated by culture and denaturing gradient gel electrophoresis (DGGE) methods. The number of cultured bacteria in flower of salt was approximately three times more than solar salt. By DGGE analysis, major microbes of DF were Maritimibacter sp., Cupriavidus sp., and unculturable bacteria, and those of DS were Cupriavidus sp., Dunalidella salina and unculturable bacteria. The results of DGGE analysis showed that major microorganisms in solar salts were composed of unidentified and unculturable bacteria and only a few microorganisms were culturable.