• 스마트미디어저널
• /
• 제12권9호
• /
• pp.45-59
• /
• 2023

Weakly supervised object localization (WSOL) is a task of localizing an object in an image using only image-level labels. Previous studies have followed the conventional class activation mapping (CAM) pipeline. However, we reveal the current CAM approach suffers from problems which cause original CAM could not capture the complete defects features. This work utilizes a convolutional neural network (CNN) pretrained on image-level labels to generate class activation maps in a multi-scale manner to highlight discriminative regions. Additionally, a vision transformer (ViT) pretrained was treated to produce multi-head attention maps as an auxiliary detector. By integrating the CNN-based CAMs and attention maps, our approach localizes defective regions without requiring bounding box or pixel-level supervision during training. We evaluate our approach on a dataset of apple images with only image-level labels of defect categories. Experiments demonstrate our proposed method aligns with several Object Detection models performance, hold a promise for improving localization.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 하계종합학술대회 논문집(1)
• /
• pp.93-96
• /
• 2000

The CMIP(Common Management Information Protocol) and SNMP(Simple Network Management Network) are the two major network management protocols, Which nee to be integrated. Since the networks need to be managed in a uniform way, the CORBA(Common Object Request Broker Architecture) NE View is to develop as a Standard in the ATM Forum For the TMN management structure to integrate these two protocols. In this paper, the function classes are defined to develop the gateway for efficient subnetwork management. The function classes are defined based on the analysis of EMS functions which are mainly network management design of the TMN structure. Also, the object models of the SMI)(Structure of Management Information) in SNMP and the GDMO(Guidelines for the Definition of Managed Objects) in CMIP are developed way direct translation and abstract translation. The integrated management system design, information model translation of EMS using classdefinition, is efficient is efficient method to interconnect CORBA/SNMP and CORBA/CMIP.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1998년도 하계학술대회 논문집 A
• /
• pp.393-395
• /
• 1998

This paper presents a modeling of railway facilities based on object-oriented software development technique for train operation simulation program. Railway network is decomposed by Line Structure Model and Signal System Model which can be composed to make the train routes and train performance calculation. A brief explanation of class design about these model is provided.

• International Journal of Control, Automation, and Systems
• /
• 제4권2호
• /
• pp.172-177
• /
• 2006

In the networked control system (NCS), the uncertain network-induced delay and nonlinear controlled object are the main problems, because they can degrade the performance of the control system and even destabilize it. In this paper, a class of uncertain and nonlinear networked control systems is discussed and its sufficient condition for the robust asymptotic stability is presented. Further, the maximum network-induced delay that insures the system stability is obtained. The Lyapunov and LMI theorems are employed to investigate the problem. The result of an illustrative example shows that the robust stability analysis is sufficient.

• Journal of Electrical Engineering and Technology
• /
• 제13권6호
• /
• pp.2468-2478
• /
• 2018

We propose the Sparse Feature Convolutional Neural Network (SFCNN) to reduce the volume of convolutional neural networks (CNNs). Despite the superior classification performance of CNNs, their enormous network volume requires high computational cost and long processing time, making real-time applications such as online-training difficult. We propose an advanced network that reduces the volume of conventional CNNs by producing a region-based sparse feature map. To produce the sparse feature map, two complementary region-based value extraction methods, cluster max extraction and local value extraction, are proposed. Cluster max is selected as the main function based on experimental results. To evaluate SFCNN, we conduct an experiment with two conventional CNNs. The network trains 59 times faster and tests 81 times faster than the VGG network, with a 1.2% loss of accuracy in multi-class classification using the Caltech101 dataset. In vehicle classification using the GTI Vehicle Image Database, the network trains 88 times faster and tests 94 times faster than the conventional CNNs, with a 0.1% loss of accuracy.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 2009년도 IWAIT
• /
• pp.709-713
• /
• 2009

Recently, a lot of research that applies data acquired from devices such as cameras and RFIDs to context aware services is being performed in the field on Life-Log and the sensor network. A variety of analytical techniques has been proposed to recognize various information from the raw data because video and audio data include a larger volume of information than other sensor data. However, manually watching a huge amount of media data again has been necessary to create supervised data for the update of a class or the addition of a new class because these techniques generally use supervised learning. Therefore, the problem was that applications were able to use only recognition function based on fixed supervised data in most cases. Then, we proposed a method of acquiring supervised data from a video sharing site where users give comments on any video scene because those sites are remarkably popular and, therefore, many comments are generated. In the first step of this method, words with a high utility value are extracted by filtering the comment about the video. Second, the set of feature data in the time series is calculated by applying functions, which extract various feature data, to media data. Finally, our learning system calculates the correlation coefficient by using the above-mentioned two kinds of data, and the correlation coefficient is stored in the DB of the system. Various other applications contain a recognition function that is used to generate collective intelligence based on Web comments, by applying this correlation coefficient to new media data. In addition, flexible recognition that adjusts to a new object becomes possible by regularly acquiring and learning both media data and comments from a video sharing site while reducing work by manual operation. As a result, recognition of not only the name of the seen object but also indirect information, e.g. the impression or the action toward the object, was enabled.

• 한국정보처리학회논문지
• /
• 제6권5호
• /
• pp.1241-1252
• /
• 1999

It is proposed in TINA, the open information telecommunication network architecture, that the definition of object group which is collection of objects provides a decrease of complex networking and a facility of object managing by service executing of application on distributed computing environment. Based on a new distributed object group model[13] we have been researched according to TINA specification, this paper proposed the object group model with the scheduler object and objects management mechanisms that can support real-time services on CORBA. To do this, we described the definition of object grouping and the requirements to suggest the object group model supporting real-time service, designed the object group structure and functional components containing in an object group using James Rumbaugh's modelling[12], and showed a class diagram of components in an object group. This paper designed IDLs of an object group manager and scheduler among the components, and finally showed the procedures of management and service interconnections between objects in the different object groups vi ETD.

• 로봇학회논문지
• /
• 제12권4호
• /
• pp.425-431
• /
• 2017

As drones gain more popularity these days, drone detection becomes more important part of the drone systems for safety, privacy, crime prevention and etc. However, existing drone detection systems are expensive and heavy so that they are only suitable for industrial or military purpose. This paper proposes a novel approach for training Convolutional Neural Networks to detect drones from images that can be used in embedded systems. Unlike previous works that consider the class probability of the image areas where the class object exists, the proposed approach takes account of all areas in the image for robust classification and object detection. Moreover, a novel loss function is proposed for the CNN to learn more effectively from limited amount of training data. The experimental results with various drone images show that the proposed approach performs efficiently in real drone detection scenarios.

• 한국항행학회논문지
• /
• 제27권2호
• /
• pp.190-196
• /
• 2023

최근 들어, 라이다 기술의 발전에 따라 정확한 거리 측정이 가능해지면서 라이다 기반의 3차원 객체 탐지 네트워크에 대한 관심이 증가하고 있다. 기존의 네트워크는 복셀화 및 다운샘플링 과정에서 공간적인 정보 손실이 발생해 부정확한 위치 추정 결과를 발생시킨다. 본 연구에서는 고수준 특징과 높은 위치 정확도를 동시에 획득하기 위해 어텐션 기반 융합 방식과 카메라-라이다 융합 시스템을 제안한다. 먼저, 그리드 기반의 3차원 객체 탐지 네트워크인 Voxel-RCNN 구조에 어텐션 방식을 도입함으로써, 다중 스케일의 희소 3차원 합성곱 특징을 효과적으로 융합하여 3차원 객체 탐지의 성능을 높인다. 다음으로, 거짓 양성을 제거하기 위해 3차원 객체 탐지 네트워크의 탐지 결과와 이미지상의 2차원 객체 탐지 결과를 결합하는 카메라-라이다 융합 시스템을 제안한다. 제안 알고리즘의 성능평가를 위해 자율주행 분야의 KITTI 데이터 세트를 이용하여 기존 알고리즘과의 비교 실험을 수행한다. 결과적으로, 차량 클래스에 대해 BEV 상의 2차원 객체 탐지와 3차원 객체 탐지 부분에서 성능 향상을 보였으며 특히 Voxel-RCNN보다 차량 Moderate 클래스에 대하여 정확도가 약 0.47% 향상되었다.

• 한국철도학회:학술대회논문집
• /
• 한국철도학회 1998년도 추계학술대회 논문집
• /
• pp.202-209
• /
• 1998

This paper presents a modelling of railway facilities and signalling system based on object-oriented software development technique to simulate multi-train movements on the complex railway network. Block and interlocking functions of signalling system is modelled using Node-Link model of railway network and signal control logic, which can be used to set the train routes and control the train movement. A brief explanation of class design about these model is provided.