Earticle

본 논문에서는 레이저스캐너만으로 이루어진 감지 시스템을 이용하여 도로 위에 있는 객체의 위치를 추정하고 분류 하는 알고리즘을 제안한다. 각각의 레이저 스캐너에서 획득한 데이터는 그리드 맵을 사용하여 데이터를 융합하였으며, 팽창 연산과 레이블링 방법을 사용하여 측정 오차를 보정하였다. 추출한 객체의 정보(길이, 폭)를 입력으로 사용한 퍼지 방법을 통해 객체를 보행자, 자전거, 차량으로 분류하였으며, 이러한 방법은 레이저스캐너로만 이루어진 감지 시스템의 정확도를 증가시켰다. 또한 본 논문에서는 실제 도로 환경에서 몇 가지 시나리오를 설정하여 실험을 하였다. 실험을 통 해 감지 시스템이 객체를 정확히 분류하는지, GPS-RTK 장비를 사용하여 획득한 위치 정보와 비교하여 객체의 위치 정 보를 정확히 추정하는지 검증하였다.

This paper proposes the on-road object detection and classification algorithm by using a detection system consisting of only laser scanners. Each sensor data acquired by the laser scanner is fused with a grid map and the measurement error and spot spaces are corrected using a labeling method and dilation operation. Fuzzy method which uses the object information (length, width) as input parameters can classify the objects such as a pedestrian, bicycle and vehicle. In this way, the accuracy of the detection system is increased. Through experiments for some scenarios in the real road environment, the performance of the proposed detection and classification system for the actual objects is demonstrated through the comparison with the actual information acquired by GPS-RTK.

최근 5년간 고속도로에서 발생한 사망 사고의 통계를 살펴보면, 고속도로 전체 사망자 중 갓길에서 발생한 사망자의 사망률이 약 3배 높은 것으로 나타났다. 이는 갓길 사고 발생 시 사고의 심각도가 매우 높다는 것을 보여주며, 갓길 차로 위반 차량을 단속하여 사고를 미연에 방지하는 것이 중요하다는 것을 시시한다. 이에 본 연구는 Faster R-CNN 기법을 활용하여 갓길 차로 위반 차량을 검출할 수 있는 방법을 제안하였다. Faster R-CNN 기법을 기반으로 차량을 탐지하고, 추가적인 판독 모듈을 구성하여 갓길 위반 여부를 판단하였다. 실험 및 평가를 위해 현실세계와 유사하게 상황을 재현할 수 있는 시뮬레이션 게임인 GTAV를 활용하였다. 이미지 형태의 학습데이터 1,800장과 평가데이터 800장을 가공 및 생성하였으며, ZFNet과 VGG16에서 Threshold 값의 변화에 따른 성능을 측정하였다. 그 결과 Threshold 0.8 기준 ZFNet 99.2%, Threshold 0.7 기준 VGG16 93.9%의 검출율을 보였고, 모델 별 평균 검출 속도는 ZFNet 0.0468 초, VGG16 0.16초를 기록하여 ZFNet의 검출율이 약 7% 정도 높았으며, 검출 속도 또한 약 3.4 배 빠름을 확인하였다. 이는 비교적 복잡하지 않은 네트워크에서도 입력 영상의 전처리 없이 빠른 속도로 갓길 차로 위반 차량의 검출이 가능함을 보여주며, 실제 영상자료 기반의 학습 데이터셋을 충분히 확보한다면 지정 차로 위반 검출에 본 알고리즘을 활용할 수 있다는 것을 시사한다.

According to the statistics about the fatal crashes that have occurred on the expressways for the last 5 years, those who died on the shoulders of the road has been as 3 times high as the others who died on the expressways. It suggests that the crashes on the shoulders of the road should be fatal, and that it would be important to prevent the traffic crashes by cracking down on the vehicles intruding the shoulders of the road. Therefore, this study proposed a method to detect a vehicle that violates the shoulder lane by using the Faster R-CNN. The vehicle was detected based on the Faster R-CNN, and an additional reading module was configured to determine whether there was a shoulder violation. For experiments and evaluations, GTAV, a simulation game that can reproduce situations similar to the real world, was used. 1,800 images of training data and 800 evaluation data were processed and generated, and the performance according to the change of the threshold value was measured in ZFNet and VGG16. As a result, the detection rate of ZFNet was 99.2% based on Threshold 0.8 and VGG16 93.9% based on Threshold 0.7, and the average detection speed for each model was 0.0468 seconds for ZFNet and 0.16 seconds for VGG16, so the detection rate of ZFNet was about 7% higher. The speed was also confirmed to be about 3.4 times faster. These results show that even in a relatively uncomplicated network, it is possible to detect a vehicle that violates the shoulder lane at a high speed without pre-processing the input image. It suggests that this algorithm can be used to detect violations of designated lanes if sufficient training datasets based on actual video data are obtained.

This study was performed to construct tree species classification map according to three information types (spectral information, texture information, and spectral and texture information) by altitude (30 m, 60 m, 90 m) using the unmanned aerial vehicle images and the object-based classification method, and to evaluate the concordance rate through field survey data. The object-based, optimal weighted values by altitude were 176 for 30 m images, 111 for 60 m images, and 108 for 90 m images in the case of Scale while 0.4/0.6, 0.5/0.5, in the case of the shape/color and compactness/smoothness respectively regardless of the altitude. The overall accuracy according to the type of information by altitude, the information on spectral and texture information was about 88% in the case of 30 m and the spectral information was about 98% and about 86% in the case of 60 m and 90 m respectively showing the highest rates. The concordance rate with the field survey data per tree species was the highest with about 92% in the case of Pinus densiflora at 30 m, about 100% in the case of Prunus sargentii Rehder tree at 60 m, and about 89% in the case of Robinia pseudoacacia L. at 90 m.

This study detected land use change using Landsat satellite image and calculated landscape index to confirm fragmentation of forest land. The land cover classification items were classified into forest land, cropland, grassland, wetlands, and settlements by referring to IPCC guidelines, and the land cover classification was used by object-based classification technique. The landscape structure analysis of land cover change was performed using five landscape index (NumP, MPS, TD, ED, AWMSI). The optimal weights for object-based classification were scale 7, Shape 0.4, Color 0.6, Compactness 0.5, Smoothness 0.5. The change of land cover was increased in the settlement area in 2009 compared to 1989 and decreased in cropland and forest land. The change of landscape index decreased NumP of cropland and forest land in 2009 compared to 1989, but TD and ED increased. This is because the settlement increased, and the cropland and forest land decreased and fragmentation progressed. In conclusion, object analysis is an effective method to confirm change detection and fragmentation, and it is expected to be used for the cause analysis of forest land fragmentation in the future.

In this study, regions infected by pine wilt disease were extracted by using object-based classification method (OB-infected region), and the characteristics of special distribution about OB-infected region were figured out. Scale 24, Shape 0.1, Color 0.9, Compactness 0.5, and Smoothness 0.5 was selected as the objected-based, optimal weighted value of OB-infected region classification. The total accuracy of classification was high with 99% and Kappa coefficient was also high with 0.97. The area of OB-infected region was approximately 90 ha, 16% of the total area. The OB-infected region in Age class V and VI was intensively distributed with 97% of the total. Also, The OB-infected region in Middle and Large DBH class was intensively distributed with 99% of the total. In terms of the topographic characteristics of OB-infected region, the damages occurred approximately 86% below the altitude of 200 m, and occurred 91% with a slope less than 10 degree. The damage occurred a lot in low hilly mountain and undulating slope. In addition, the accessibility to road and residential area from OB-infected region was less than 300 m in large part. Overall, it was figured out that artificial effect is stronger than natural effect with regard to the spread of pine wilt disease.

Along with the current rapid development of technology, object classification is being researched, developed, and applied to security systems, autonomous driving, and other applications. A common technique is to use vision cameras to collect data of objects in the surrounding environment. Along with many other methods, LiDAR sensors are being used to collect data in space to detect and classify objects. By using the LiDAR sensors, some disadvantages of image sensors with the negative influence on the image quality by weather and light condition will be covered. In this study, a volumetric image descriptor in 3D shape is developed to handle 3D object data in the urban environment obtained from LiDAR sensors, and convert it into image data before using deep learning algorithms in the process of object classification. The study showed the potential possibility of the proposal and its further application.

본 논문에서는 상황 변화 환경에서 적응적인 객체 인식을 위한 계층적 트리 구조를 이용한 조명 온톨로지 분류에 대한 방식을 제안한다. 본 논문에서는 상황이 불변하는 환경에서 동작하는 개발된 많은 시스템을 찾아냈고, 상황에 맞는 감지를 위한 새로운 개념의 트리 구조를 이용한 온톨로지를 도입하였다. 조명의 영향이 상황 인지 인식 시스템을 아주 설계하기 어려운 시스템으로 만들기 때문에 본 논문에서는 트리 구조의 온톨로지를 사용하여 이러한 상황 변화 시스템을 설계하는데 더 중점을 두었다. 온톨로지는 일반적으로 사람들이 특정 분야의 것들에 대해 생각하는 방법의 추상적 모델에서 전형적으로 캡처된 한 분야의 개념화의 명시적 사양으로 정의 할 수 있다. 인간은 기본 원칙과 환경을 이해하고 설명하기 위해 온톨로지를 생성한다. 본 연구에서는 상황 온톨로지, 상황 모델링, 상황 적응 및 조명 기준에 따라 트리 구조 온톨로지를 설계하는 상황 분류를 제안했다. 조명 온톨로지의 적당한 영역을 선택한 후, 그 영역에서 더 나은 성능을 생산하는 동작의 한 집합을 선택하는데 있어서 장점을 얻었다. 본 논문에서는 역동적인 변화 환경에서 객체 인식의 영역에서 이러한 개념을 이용하여 폭 넓은 실험을 수행하였으며 제안하는 기본 개념에 대해 수행할 수 있는 많은 성공을 얻었다.

This paper proposes a ontology of tree structure approach for adaptive object recognition in a situation-variant environment. In this paper, we introduce a new concept, ontology of tree structure ontology, for context sensitivity, as we found that many developed systems work in a context-invariant environment. Due to the effects of illumination on a supreme obstinate designing context-sensitive recognition system, we have focused on designing such a context-variant system using ontology of tree structure. Ontology can be defined as an explicit specification of conceptualization of a domain typically captured in an abstract model of how people think about things in the domain. People produce ontologies to understand and explain underlying principles and environmental factors. In this research, we have proposed context ontology, context modeling, context adaptation, and context categorization to design ontology of tree structure based on illumination criteria. After selecting the proper light-ontology domain, we benefit from selecting a set of actions that produces better performance on that domain. We have carried out extensive experiments on these concepts in the area of object recognition in a dynamic changing environment, and we have achieved enormous success, which will enable us to proceed on our basic concepts.

Quiet often, an organization tries to grapple with inconsistent and partial information to generate relevant information to support decision making and action. As such, an organization scans the environment interprets scanned data, executes actions, and learns from feedback of actions, which boils down to computational interpretations and learning in terms of machine learning, statistics, and database. The ExOM proposed in this paper is geared to facilitate such knowledge discovery found in large databases in a most flexible manner. It supports a broad range of learning and classification styles and integrates them with traditional database functions. The learning and classification components of the ExOM are tightly integrated so that learning and classification of objects is less burdensome to ordinary users. A brief sketch of a strategy as to the expressiveness of terminological language is followed by a description of prototype implementation of the learning and classification components of the ExOM.

본 논문은 서베일런스 네트워크에서 이동하는 객체 추적 시 영상 데이터의 전송량을 감소시키는 신경망 계산 시간의 단축 알고리즘을 제안한다. 객체 검출은 디지털화 연속된 영상으로부터 객체 존재 유무를 판단하고, 객체가 존재할 경우 영상 내 객체의 위치, 방향, 크기 등을 알아내는 기술로 정의된다. 그러나 영상 내의 객체는 위치, 크기, 빛의 방향 및 밝기, 장애물 등의 환경적 변화로 인해 객체 모양이 다양해지므로 정확하고 빠른 검출이 어렵다. 따라서 본 논문에서는 신경망을 사용하여 몇 가지 환경적 조건을 극복한 정확하고 빠른 객체 검출 방법을 제안한다. 검색 영역의 축소는 영상 내 색상 영역의 분할과 차영상을 이용하였고, 주성분 분석을 통해 신경망의 입력 벡터를 축소시킴으로써 신경망 수행 시간과 학습 시간을 단축시켰다. 실시간으로 입력되는 동영상에서 모두 실험하였으며, 색상 영역의 분할을 사용할 경우 입력 영상의 칼라 설정의 유무에 따른 검출 성공률의 차를 보였다. 실험 결과에서 보면 제안하는 방법으로써 객체의 움직임을 탐지하였을 때 기존의 방법보다 30% 정도 더 높은 인식 성능을 보여준다.

This paper proposes algorithm to reduce the computing time in a neural network that reduces transmission of data for tracking mobile objects in surveillance networks in terms of detection and communication load. Object Detection can be defined as follows : Given image sequence, which can forom a digitalized image, the goal of object detection is to determine whether or not there is any object in the image, and if present, returns its location, direction, size, and so on. But object in an given image is considerably difficult because location, size, light conditions, obstacle and so on change the overall appearance of objects, thereby making it difficult to detect them rapidly and exactly. Therefore, this paper proposes fast and exact object detection which overcomes some restrictions by using neural network. Proposed system can be object detection irrelevant to obstacle, background and pose rapidly. And neural network calculation time is decreased by reducing input vector size of neural network. Principle Component Analysis can reduce the dimension of data. In the video input in real time from a CCTV was experimented and in case of color segment, the result shows different success rate depending on camera settings. Experimental results show proposed method attains 30% higher recognition performance than the conventional method.

생활 폐기물 쓰레기는 기후 변화, 자원 부족, 환경 오염을 불러오는 대표적인 문제로서, 이러한 문제를 해결하기 위 해 지능적으로 쓰레기를 분류하는 방식을 연구하였고, 전통적인 분류 알고리즘부터 기계학습, 신경망에 이르기까지 많 은 연구가 진행되고 있다. 그러나, 다양한 환경과 조건에서 쓰레기를 분류하기에는 여전히 데이터셋이 부족하고, 신경망 네트워크 구성 복잡도가 증가하며, 성능 측면에서도 실생활에 적용하기에 아직 미흡하다. 따라서 본 논문에서는 신속한 분류와 정확도 향상을 위해 i-YOLOX를 제안하고, 네트워크 매개변수, 검출속도, 정확도 등을 평가한다. 이를 위해 17 개의 폐기물 범주를 포함하는 10,000개의 가정용 쓰레기 대상 샘플로 데이터 세트를 구성하고, YOLOX 구조에 Involution 채널 컨볼루션 연산자와 CBAM(Convolution Branch Attention Module)을 도입하여 i-YOLOX를 구성하고, 기존의 YOLO 구조와 성능을 비교한다. 실험 결과 복잡한 장면에서 쓰레기 객체 검출 속도와 정확도가 기존의 신경망 에 비해 향상되어, 제안한 i-YOLOX 구조가 생활 폐기물 다중 객체 검출과 분류에 효과적임을 확인하였다.

In addressing the prominent issues of climate change, resource scarcity, and environmental pollution associated with ho usehold waste, extensive research has been conducted on intelligent waste classification methods. These efforts range fro m traditional classification algorithms to machine learning and neural networks. However, challenges persist in effectively classifying waste in diverse environments and conditions due to insufficient datasets, increased complexity in neural netwo rk architectures, and performance limitations for real-world applications. Therefore, this paper proposes i-YOLOX as a sol ution for rapid classification and improved accuracy. The proposed model is evaluated based on network parameters, detect ion speed, and accuracy. To achieve this, a dataset comprising 10,000 samples of household waste, spanning 17 waste cate gories, is created. The i-YOLOX architecture is constructed by introducing the Involution channel convolution operator an d the Convolution Branch Attention Module (CBAM) into the YOLOX structure. A comparative analysis is conducted with the performance of the existing YOLO architecture. Experimental results demonstrate that i-YOLOX enhances the detectio n speed and accuracy of waste objects in complex scenes compared to conventional neural networks. This confirms the eff ectiveness of the proposed i-YOLOX architecture in the detection and classification of multiple household waste objects.

실험실에서 발생하는 화재 사고는 인명과 재산 피해 위험이 커 신속한 탐지와 체계적인 대응이 요구된다. 본 논문에서는 실험실에서 발생할 수 있는 다양한 화재 상황에 효율적으로 대처하기 위하여 4종류(일반, 유류, 전기, 금속) 화재 시나리오를 포함하는 이미지 데이터 셋을 구축하고, 객체 검출(YOLOv5) 기반 화재 특징 추출과 트리 기반 의사결정(Random Forest) 모형을 결합한 이미지 기반 실험실 화재 유형 분류 모델을 제시하고 평가한다.

공공장소에서의 유기물은 의도적 공공테러를 목적으로 폭발물이나 화학물질 등을 포함할 수 있기 때문에 일단 가능한 위험물로 반드시 다루어져야 한다. 공항이나 기차역과 같은 대형 공공장소에서는 전체 영역을 감시하는 모든 모니터를 점검할 보안 인력을 유지하는데 있어서 비용적 측면의 한계가 있게 마련이다. 이것이 비전 기술에 기반한 위험 유기물의 자동 검사 시스템을 개발하여야 하는 기본적 동기이다. 이 연구에서는 잘 알려진 DBE 기법을 적용하여 배경 이미지를 안정적으로 추출하는 것을 보이며, HOG 알고리즘을 적용하여 물체 분류에 있어서 사람과 물건을 구분하는 기능을 구현하였다. 제안된 시스템의 유효성을 보이기 위하여 감시 지역의 한 실내 환경에 대해 금지 구역 침범을 탐지하고 유기물에 대한 경보를 발생하는 실험을 수행하였다.

Abandoned objects should be treated as possibly dangerous things for public areas until they turn out to be safe because explosive material or chemical substance is intentionally contained in them for public terrors. For large public areas such as airports or train stations, there are limits in man-power for security staffs to check all the monitors for covering the entire area under surveillance. This is the basic motivation of developing the automatic detection system for dangerous abandoned objects based on vision technology. In this research, well-known DBE is applied to stably extract background images and the HOG algorithm is adapted to discriminate between human and stuff for object classification. To show the effectiveness of the proposed system, experiments are carried out in detecting intrusion for a forbidden area and alarming for abandoned objects in a room under surveillance.

Security checks at airport are of importance to any safe flight. Traditional method for security check is to check the luggage manually. However, this method needs lots of human labor and time. It is desired to automatically check x-ray image of the luggage with computer vision. In this paper, the illegal object classification system is implemented. First, we introduce a computer vision based x-ray inspection system for Illegal object classification. Then we propose a method by combining Taruma feature based on Contourlet transform and histogram. Finally, we apply the random forests classifier to classify these features from the illegal objects. Experimental results show that the proposed method can effectively distinguish different kinds of illegal objects.

The Classifier Integration Model (CIM) with Convolutional Neural Networks (CNNs) as its local classifiers is applied to an object classification task for vehicle safety system in this paper. The Convolutional Neural Networks adopted in this paper has a very unique advantage when compared with conventional neural network models because CNNs do not require any feature extraction procedure prior to the classification process while other existing classification methods require rather very complex feature extraction process. Several CNN models are first designed as local classifiers and these models are then combined to make a decision in Classifier Integration Model for our classification task. Experiments on real data sets obtained for our experiments show that the CNN-based CIM scheme gives a promising performance in terms of training speed and classification accuracy.

Public surveillance monitoring is rapidly finding its way into Intelligent Surveillance System. Street crime is increasing in recent years, which has demanded more reliable and intelligent public surveillance system. In this paper, the ability and the accuracy of an Adaptive Neuro-Fuzzy Inference System (ANFIS) was investigated for the classification of moving objects for street scene applications. The goal of this paper is to classify the moving objects prior to its communal attributes that emphasize on three major processes which are object detection, discriminative feature extraction, and classification of the target. The intended surveillance application would focus on street scene, therefore the target classes of interest are pedestrian, motorcyclist, and car. The adaptive network based on Neuro-fuzzy was independently developed for three output parameters, each of which constitute of three inputs and 27 Sugeno-rules. Extensive experimentation on significant features has been performed and the evaluation performance analysis has been quantitatively conducted on three street scene dataset, which differ in terms of background complexity. Experimental results over a public dataset and our own dataset demonstrate that the proposed technique achieves the performance of 93.1% correct classification for street scene with moving objects, with compared to the solely approaches of neural network or fuzzy.

In this paper, we propose a novel, fast model for detecting fire flame and smoke using object detection and 3D classification, referred to as FastFireDet3D. This model uses NanoDet to quickly identify potential areas representing fire and smoke, followed by a novel 3D classification model based on a spatio-temporal convolutional neural network (STCNN). This two-step process allows for efficient and accurate detection. The average processing time for FastFireDet3D is approximately 40-90ms when run on a CPU, and it achieves an accuracy improvement of 3.45% over traditional Convolutional 3D (C3D) models.

In this paper, we proposed the object classification method using genetic and dynamic random forest consisting of optimal combination of unit tree. The random forest can ensure good generalization performance in combination of large amount of trees by assigning the randomization to the training samples and feature selection, etc. allocated to the decision tree as an ensemble classification model which combines with the unit decision tree based on the bagging. However, the random forest is composed of unit trees randomly, so it can show the excellent classification performance only when the sufficient amounts of trees are combined. There is no quantitative measurement method for the number of trees, and there is no choice but to repeat random tree structure continuously. The proposed algorithm is composed of random forest with a combination of optimal tree while maintaining the generalization performance of random forest. To achieve this, the problem of improving the classification performance was assigned to the optimization problem which found the optimal tree combination. For this end, the genetic algorithm methodology was applied. As a result of experiment, we had found out that the proposed algorithm could improve about 3~5% of classification performance in specific cases like common database and self infrared database compare with the existing random forest. In addition, we had shown that the optimal tree combination was decided at 55~60% level from the maximum trees.

본 논문에서는 텍스쳐와 컬러 정보를 기반으로 비디오 감시를 위한 빠른 물체 분류 방법을 제안한다. 영상들로부터 SURF와 색 히스토그램의 국부적 패치들을 추출하여 그들의 장점을 이용한다. SURF는 명암 내용 정보를 제공하고 색 정보는 패치에 대한 특이성을 증강시킨다. SURF의 빠른 계산뿐만 아니라 객체의 색 정보를 활용한다. 국부적 특징을 이용하여 관심 영역 혹은 영상의 전역적 서술자를 생성하기 위해 Bag of Word 모델을 이용하고, 전역적 서술자를 분류하기 위해 Na$\ddot{i}$ve Bayes 모델을 이용한다. 또한 본 논문에서는 판별적인 기술자인 SIFT도 성능 분석한다. 네 종류의 객체에 대한 실험결과 95.75%의 인식률을 보였다.

In this paper, we propose a fast object classification method based on texture and color information for video surveillance. We take the advantage of local patches by extracting SURF and color histogram from images. SURF gives intensity content information and color information strengthens distinctiveness by providing links to patch content. We achieve the advantages of fast computation of SURF as well as color cues of objects. We use Bag of Word models to generate global descriptors of a region of interest (ROI) or an image using the local features, and Na$\ddot{i}$ve Bayes model for classifying the global descriptor. In this paper, we also investigate discriminative descriptor named Scale Invariant Feature Transform (SIFT). Our experiment result for 4 classes of the objects shows 95.75% of classification rate.