동영상에서 보행자 추적을 위한 변형된 HOG 특징 추출에 관한 연구

김회준,박영수,김기봉,이상훈 한국융합학회 2019 한국융합학회논문지 Vol.10 No.3

In this paper, we proposed extracting modified Histogram of Oriented Gradients (HOG) features using background removal when tracking pedestrians in real time. HOG feature extraction has a problem of slow processing speed due to large computation amount. Background removal has been studied to improve computation reductions and tracking rate. Area removal was carried out using S and V channels in HSV color space to reduce feature extraction in unnecessary areas. The average S and V channels of the video were removed and the input video was totally dark, so that the object tracking may fail. Histogram equalization was performed to prevent this case. HOG features extracted from the removed region are reduced, and processing speed and tracking rates were improved by extracting clear HOG features. In this experiment, we experimented with videos with a large number of pedestrians or one pedestrian, complicated videos with backgrounds, and videos with severe tremors. Compared with the existing HOG-SVM method, the proposed method improved the processing speed by 41.84% and the error rate was reduced by 52.29%. 본 논문에서는 실시간으로 보행자를 추적할 때 배경 제거를 이용하여 변형된 HOG(Histogram of Oriented Gradients) 특징 추출을 제안하였다. 기존의 HOG 특징 추출은 연산량이 많아 추적 속도가 느린 문제가 있다. 배경 제거를 통해 연산량 감소와 추적률을 향상시키기 위해 연구하였다. 불필요한 영역에서의 특징 추출을 감소시키기 위해 HSV 색공간에서 S 채널과 V 채널을 이용하여 영역 제거를 진행하였다. 영상의 평균 S 채널과 V 채널로 배경 제거 후 입력 영상이 전체적으로 어두워 객체 추적에 실패하는 경우가 있다. 이러한 경우를 방지하기 위해 히스토그램 평활화를 하였다. 제거된 영역에서 추출되는 HOG 특징이 감소되고, 객체에서는 명확한 HOG 특징이 추출되어 객체 추적 속도와 추적률이 향상되었다. 본 실험에서는 다수의 보행자나 한명의 보행자가 존재하는 영상, 배경이 복잡한 영상, 흔들림이 심한 영상을 가지고 실험하였다. 제안하는 방법은 기존의 HOG-SVM 방법과 비교하여 처리 속도는 약 41.84% 향상되었으며 오 추적률은 약 52.29% 감소되어 개선된 추적률을 보였다.

양돈 수직계열주체의 지원과 통제가 계열농가의 재계약 의도에 미치는 영향 연구

최형규 ( Choi Hyung Kyu ),김동환 ( Kim Dong Hwan ) 한국농업정책학회 2014 농업경영정책연구 Vol.41 No.1

This study aims to examine effects of support and control of vertical integrators on the recontracting intention of contractees in hog/pork industry. The 202 hog farms of 11 pork brands were surveyed. The collected data were analyzed by a Structural Equation Model (SEM) using AMOS ver. 16.0 to investigate relationship among the variables in a comprehensive way. The results of empirical analysis show that support of hog integrators had a significantly positive(+) effect on the satisfaction of the hog farms, and control of hog integrators had a significantly positive(+) effect on conflict between integrators and the hog farms. This study also shows that satisfaction of the hog farms has a significantly positive(+) effect on recontracting intention of the hog farms and that conflict between integrators and the hog farms has a significantly negative(-) effect on recontracting intention of the hog farms. These empirical results imply that the hog/pork integrators have to strengthen support activities, such as high quality sow, management consulting, operational finance, and etc. They also have to develop elaborate control activities, such as designation of feed, management program, monitoring and etc. To decrease conflict, it is recommended that trust relation-ship between two parties should be established by creating communication channels and conciliation procedures.

곡률과 HOG에 의한 연속 방법에 기반한 아다부스트 알고리즘을 이용한 보행자 인식

이영화(Yeunghak Lee),고주영(Jooyoung Ko),석정희(Jung-Hee Suk),노태문(Tae Moon Roh),심재창(Jaechang Shim) 한국정보과학회 2010 정보과학회 컴퓨팅의 실제 논문지 Vol.16 No.6

본 논문은 2단계 연속(cascade) 방법을 이용한 향상된 보행자/비보행자 인식 알고리즘을 제안한다. 인식을 위한 분류기로는 약한 분류기를 강한 분류기로 만드는 아다부스트 알고리즘을 적용하였다. 먼저 두 가지 특징벡터를 추출 한다: (i) 기존의 기울기 히스토그램(HOG) 특성과 (ii) 한 점이 가지는 곡률특성 네 가지를 이용한 곡률-HOG를 제안하고 이용하였다. 그 다음 훈련 영상을 통하여 두 가지의 특징 벡터에 대해 약한 분류기로부터 강한 분류기를 얻었으며, 인식은 입력 영상으로부터 하나의 특징을 선택하여 이미 만들어진 강한 분류기를 통하여 1차적인 인식과 오인식을 실시하며, 오인식된 영상에 대해 2차적인 특징을 투입하여 이에 해당하는 강한 분류기를 통하여 2단계 아다부스트 알고리즘을 적용하여 최종적인 인식결과를 얻는다. 두 가지의 서로 다른 특성 벡터를 이용하여 연속 방법에 의한 2단계 아다부스트 알고리즘을 적용한 결과 기존의 실험 방법보다 더 정확한 인식 결과를 얻을 수 있었다. In this paper, we suggest an advanced algorithm, to recognize pedestrian/non-pedestrian using second-stage cascade method, which applies Adaboost algorithm to make a strong classification from weak classifications. First, we extract two feature vectors: (i) Histogram of Oriented Gradient (HOG) which includes gradient information and differential magnitude; (ii) Curvature-HOG which is based on four different curvature features per pixel. And then, a strong classification needs to be obtained from weak classifications for composite recognition method using both HOG and curvature-HOG. In the proposed method, we use one feature vector and one strong classification for the first stage of recognition. For the recognition-failed image, the other feature and strong classification will be used for the second stage of recognition. Based on our experiment, the proposed algorithm shows higher recognition rate compared to the traditional method.

Comparison of Airborne Bacterial Communities from a Hog Farm and Spray Field

( Ann M. Arfken ),( Bong Keun Song ),( Jung Suk Sung ) 한국미생물 · 생명공학회 2015 Journal of microbiology and biotechnology Vol.25 No.5

Airborne bacteria from hog farms may have detrimental impacts on human health, particularly in terms of antibiotic resistance and pathogen zoonosis. Despite human health risks, very little is known about the composition and diversity of airborne bacteria from hog farms and hog-related spray fields. We used pyrosequencing analysis of 16S rRNA genes to compare airborne bacterial communities in a North Carolina hog farm and lagoon spray field. In addition, we isolated and identified antibiotic-resistant bacteria from both air samples. Based on 16S rRNA gene pyrosequence analysis, Actinobacteria, Bacteroidetes, Firmicutes, and Proteobacteria were the dominant phyla in airborne bacterial communities from both hog farm and spray field sites. Within the Firmicutes genera, Clostridium spp. were more abundant in the hog farm, whereas Staphylococcus spp. were higher in the spray field. The presence of opportunitic pathogens, including several Staphylococcus species and Propionibacterium acnes, was detected in both bioaerosol communities based on phylogenetic analysis. The isolation and identification of antibiotic-resistant bacteria from air samples also showed similar results with dominance of Actinobacteria and Proteobacteria in both hog farm and spray field air. Thus, the existence of opportunistic pathogens and antibiotic resistant bacteria in airborne communities evidences potential health risks to farmers and other residents from swine bioaerosol exposure.

HOG기반 RBFNN을 이용한 상반신 검출 시스템의 설계

김선환(Sun-Hwan Kim),오성권(Sung-Kwun Oh),김진율(Jin-Yul Kim) 한국지능시스템학회 2016 한국지능시스템학회논문지 Vol.26 No.4

최근 감시와 보안을 목적으로 활발하게 CCTV가 설치되고 있고, 지능형 감시시스템은 영상에서 객체의 검출 및 감시 등으로 광범위하게 응용되고 있다. 본 연구에서는 지능형 영상 감시 시스템에서 HOG 특징과 FCM 기반의 RBFNN 분류기를 이용한 상반신 검출 방법을 제안한다. HOG는 보행자를 검출하기 위해 기존에 제안되었던 특징으로 본 논문에서는 이를 사용해 상반신의 고유한 기울기를 학습하였다. HOG 특징은 입력 이미지의 크기에 비례하는 고차원의 특징 벡터로 기울기를 표현하기 때문에 RBFNN분류기의 입력데이터로 쓰려면 차원 축소가 필요하다. 이를 위해 PCA 알고리즘을 RBFNN 분류기 앞에 적용하여 HOG 특징의 차원을 저차원으로 축소하였다. 컴퓨터 실험에서는 미리 분류된 상반신 영상과 사람이 아닌 영상을 통해 분류기를 훈련시킨 후 테스트 영상과 동영상을 이용하여 제안된 상반신 검출 방법의 성능을 평가하였다 Recently, CCTV cameras are emplaced actively to reinforce security and intelligent surveillance systems have been under development for detecting and monitoring of the objects in the video. In this study, we propose a method for detection of upper body in intelligent surveillance system using FCM-based RBFNN classifier realized with the aid of HOG features. Firstly, HOG features that have been originally proposed to detect the pedestrian are adopted to train the unique gradient features about upper body. However, HOG features typically exhibit a very high dimension of which is proportional to the size of the input image, it is necessary to reduce the dimension of inputs of the RBFNN classifier. Thus the well-known PCA algorithm is applied prior to the RBFNN classification step. In the computer simulation experiments, the RBFNN classifier was trained using pre-classified upper body images and non-person images and then the performance of the proposed classifier for upper body detection is evaluated by using test images and video sequences.

Regulation of MAP kinase Hog1 by calmodulin during hyperosmotic stress

Kim, J.,Oh, J.,Sung, G.H. Elsevier Biomedical Press 2016 Biochimica et biophysica acta, Molecular cell rese Vol.1863 No.11

Mitogen-activated protein kinase (Hog1 in yeast and ortholog p38 in human cells) plays a critical role in the signal transduction pathway that is rapidly activated under multiple stress conditions. Environmental stress stimuli such as hyperosmotic stress cause changes in cellular ATP metabolism required for hyperosmotic stress tolerance. Furthermore, hyperosmotic stress induces rapid Ca<SUP>2+</SUP> signals in eukaryotic cells. These Ca<SUP>2+</SUP> signals can be decoded by Ca<SUP>2+</SUP> sensor calmodulin (CaM). By using genetic and biochemical approaches, we demonstrate that Hog1 is a novel CaM-binding protein, and that CaM-binding to Hog1 is involved in the mediation of the hyperosmotic stress signaling pathway. In addition, we show that p38α, a human ortholog of Hog1, interacts with CaM, suggesting that the CaM-binding feature of Hog1/p38α is evolutionarily conserved in eukaryotic cells. Hog1 is likely involved in cellular ATP regulation through CaM signaling during hyperosmotic stress. Therefore, this work suggests that Hog1 plays an important role in connecting CaM signaling with the hyperosmotic stress pathway by directly interacting with CaM in Saccharomyces cerevisiae.

Construction of an Efficient Mutant Strain of Trichosporonoides oedocephalis with HOG1 Gene Deletion for Production of Erythritol

( Liangzhi Li ),( Tianyi Yang ),( Weiqiang Guo ),( Xin Ju ),( Cuiying Hu ),( Bingyu Tang ),( Jiaolong Fu ),( Jingsheng Gu ),( Haiyang Zhang ) 한국미생물 · 생명공학회 2016 Journal of microbiology and biotechnology Vol.26 No.4

The mitogen-activated protein kinase HOG1 (high-osmolarity glycerol response pathway) plays a crucial role in the response of yeast to hyperosmotic shock. Trichosporonoides oedocephalis produces large amounts of polyols (e.g., erythritol and glycerol) in a culture medium. However, the effects of HOG1 gene knockout and environmental stress on the production of these polyols have not yet been studied. In this study, a To-HOG1 null mutation was constructed in T. oedocephalis using the loxP-Kan-loxP/Cre system as replacement of the targeted genes, and the resultant mutants showed much smaller colonies than the wild-type controls. Interestingly, compared with the wild-type strains, the results of shake-flask culture showed that To-HOG1 null mutation increased erythritol production by 1.44-fold while decreasing glycerol production by 71.23%. In addition, this study investigated the effects of citric acid stress on the T. oedocephalis HOG1 null mutants and the wild-type strain. When the supplementation of citric acid in the fermentation medium was controlled at 0.3% (w/v), the concentration of erythritol produced from the wild-type and To-HOG1 knockout mutant strains improved by 18.21% and 21.65%, respectively.

HOG 특징 및 영상분할을 이용한 부스팅분류 기반 자동차 검출 기법

최미순(Mi Soon Choi),이정환(Jeong Hwan Lee),노태문(Tae Moon Roh),심재창(Jaechang Shim) 한국정보과학회 2010 정보과학회 컴퓨팅의 실제 논문지 Vol.16 No.10

본 논문에서는 HOG 특징벡터와 영상분할을 이용한 부스팅 분류기반의 자동차영역 검출 알고리즘의 연구에 대해서 기술한다. 입력된 영상으로부터 차량을 검출하기위해 먼저 분할 후 합병(split-merge) 방법을 적용하여 영상을 분할한다. 그리고 가장 큰 두 영역을 검색 영역에서 제외하여 처리 속도를 향상 시킨다. 각 영역에 대해 HOG(histogram of oriented gradient) 특징을 추출한다. 분류기는 두 개의 모집단을 분류하는데 많이 사용되고 있는 AdaBoost 방법을 사용한다. 제안방법의 성능 평가를 위해 537개의 영상을 사용하여 분류기를 학습하였으며, 또한 학습에 사용하지 않은 비학습영상 500개를 사용하여 인식률을 구하였다. 실험결과 비학습영상에 대해 98.34%의 인식률을 얻었다. 결론적으로 제안된 방법이 지능형 자동차 제어 시스템에서 차량의 위치를 찾는 방법으로 활용될 수 있다. In this paper, we describe a study of a vehicle detection method based on a Boosting Classifier which uses Histogram of Oriented Gradient (HOG) features and Image Segmentation techniques. An input image is segmented by means of a split and merge algorithm. Then, the two largest segmented regions are removed in order to reduce the search region and speed up processing time. The HOG features are then calculated for each pixel in the search region. In order to detect the vehicle region we used the AdaBoost (adaptive boost) method, which is well known for classifying samples with two classes. To evaluate the performance of the proposed method, 537 training images were used to train and learn the classifier, followed by 500 non-training images to provide the recognition rate. From these experiments we were able to detect the proper image 98.34% of the time for the 500 non-training images. In conclusion, the proposed method can be used for detecting the location of a vehicle in an intelligent vehicle control system.

A Real-time Pedestrian Detection based on AGMM and HOG for Embedded Surveillance

Nguyen, Thanh Binh,Nguyen, Van Tuan,Chung, Sun-Tae Korea Multimedia Society 2015 멀티미디어학회논문지 Vol.18 No.11

Pedestrian detection (PD) is an essential task in various applications and sliding window-based methods utilizing HOG (Histogram of Oriented Gradients) or HOG-like descriptors have been shown to be very effective for accurate PD. However, due to exhaustive search across images, PD methods based on sliding window usually require heavy computational time. In this paper, we propose a real-time PD method for embedded visual surveillance with fixed backgrounds. The proposed PD method employs HOG descriptors as many PD methods does, but utilizes selective search so that it can save processing time significantly. The proposed selective search is guided by restricting searching to candidate regions extracted from Adaptive Gaussian Mixture Model (AGMM)-based background subtraction technique. Moreover, approximate computation of HOG descriptor and implementation in fixed-point arithmetic mode contributes to reduction of processing time further. Possible accuracy degradation due to approximate computation is compensated by applying an appropriate one among three offline trained SVM classifiers according to sizes of candidate regions. The experimental results show that the proposed PD method significantly improves processing speed without noticeable accuracy degradation compared to the original HOG-based PD and HOG with cascade SVM so that it is a suitable real-time PD implementation for embedded surveillance systems.

A Ferroxidase, Cfo1, Regulates Diverse Environmental Stress Responses of Cryptococcus neoformans through the HOG Pathway

( Kyung Tae Lee ),( Jang Won Lee ),( Dohyun Lee ),( Won Hee Jung ),( Yong Sun Bahn ) 한국균학회 2014 韓國菌學會誌 Vol.42 No.2

The iron uptake and utilization pathways play a critical role in allowing human pathogens, including Cryptococcus neoformans, the causative agent of fatal meningoencephalitis, to survive within the mammalian body by competing with the host for iron. Here we show that the iron regulon is also required for diverse environmental stress responses and that in C. neoformans, it is regulated by the high-osmolarity glycerol response (HOG) pathway. Between CFO1 and CFO2, two ferroxidase genes in the iron regulon, CFO1 but not CFO2 was induced during oxidative and osmotic stress. Interestingly, we found that the HOG pathway repressed basal expression of both CFO1 and CFO2. Furthermore, when the HOG pathway was blocked, CFO2 also responded to oxidative and osmotic stress and the response of CFO1 was increased. We also established that CFO1 plays a major role in responding and adapting to diverse environmental stresses, including oxidative and genotoxic damage, osmotic fluctuations, heavy metal stress, and stress induced by cell membrane destabilizers. Therefore, our findings indicate that in C. neoformans, the iron uptake and utilization pathways are not only required for iron acquisition and survival, but also play a significant role in the environmental stress response through crosstalk with the HOG pathway.