• The KIPS Transactions:PartB
• /
• v.17B no.5
• /
• pp.375-380
• /
• 2010

Most of previous algorithms of object boundary extraction have been studied for extracting the boundary of single object. However, multiple objects are much common in the real image. The proposed algorithm of extracting the boundary of each of multiple objects has two steps. In the first step, we propose the fast method using the outer and inner products; the initial contour including multiple objects is split and connected and each of new contours includes only one object. In the second step, an improved active contour model is studied to extract the boundary of each object included each of contours. Experimental results with various test images have shown that our algorithm produces much better results than the previous algorithms.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.5
• /
• pp.168-176
• /
• 2014

Detecting moving objects from a video sequence is a fundamental and critical task in video surveillance, traffic monitoring and analysis, and human detection and tracking. It is very difficult to detect moving objects in a video sequence degraded by the environmental factor such as fog. In particular, the color of an object become similar to the neighbor and it reduces the saturation, thus making it very difficult to distinguish the object from the background. For such a reason, it is shown that the performance and reliability of object detection and tracking are poor in the foggy weather. In this paper, we propose a novel method to improve the performance of object detection, combining a haze removal algorithm and a local histogram-based object tracking method. For the quantitative evaluation of the proposed system, information retrieval measurements, recall and precision, are used to quantify how well the performance is improved before and after the haze removal. As a result, the visibility of the image is enhanced and the performance of objects detection is improved.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.06a
• /
• pp.244-245
• /
• 2022

최근에는 우수한 성능의 딥러닝 기술을 활용한 장비와 프로그램이 개발되고 있으나 기술의 특성상 모든 환경에서 우수한 성능을 보여주지 못하고 고 사양의 서버와 같은 환경에서의 성능만을 보장하고 있다. 따라서 이에 대한 개선으로 엣지 디바이스 독립적으로 혹은 클라우드 의존과 인터넷 연결을 최소화 할 수 있는 엣지 컴퓨팅 기술이 제안되고 있으며 경량 내장형 시스템에 적합한 인공지능 기술의 개발이 필요하다. 본 논문에서는 객체검출 모델을 적은 연산과 효율적인 구조로 설계하고 생성된 모델을 임베디드 보드에서 원활하게 실행할 수 있도록 중립 모델로 변환하고 경량화 하는 방법에 대해 소개한다. Qualcomm snapdragon 프로세서가 갖춰진 임베디드 보드를 목표로 하였고 편의를 위해 SNPE(snapdragon neural processing engine) SDK를 이용하여 실험을 진행하였다. 실험 결과 변환된 중립모델이 기존 모델과 비교하여 압축된 모델 크기 대비 미미한 성능 저하가 발생함을 확인할 수 있었다.

• Journal of Convergence for Information Technology
• /
• v.11 no.9
• /
• pp.28-34
• /
• 2021

In the case of object detection using deep learning, both accuracy and real-time are required. However, it is difficult to use a deep learning model that processes a large amount of data in a limited resource environment. To solve this problem, this paper proposes an object detection model for small embedded devices. Unlike the general detection model, the model size was minimized by using a structure in which the pre-trained feature extractor was removed. The structure of the model was designed by repeatedly stacking lightweight convolution blocks. In addition, the number of region proposals is greatly reduced to reduce detection overhead. The proposed model was trained and evaluated using the public dataset PASCAL VOC. For quantitative evaluation of the model, detection performance was measured with average precision used in the detection field. And the detection speed was measured in a Raspberry Pi similar to an actual embedded device. Through the experiment, we achieved improved accuracy and faster reasoning speed compared to the existing detection method.

• Journal of the Korea Society of Computer and Information
• /
• v.12 no.2 s.46
• /
• pp.71-76
• /
• 2007

Object detection process which makes decision on the existence of meaningful objects in a given image is a crucial part of image recognition in computer vision system. Hausdorff distance metric has been used in object detection and shows good results in applications such as face recognition. It defines the dissimilarity between two sets of points and is used to find the object that is most similar to the given model. This paper proposes a Hausdorff distance based detection method that uses directional information of points to improve detection accuracy when the sets of points are derived from edge extraction as is in usual cases. In this method, elastic energy needed to make two directional points coincident is used as a measure of similarity.

• Journal of Digital Convergence
• /
• v.13 no.9
• /
• pp.183-189
• /
• 2015

In this paper, we propose an object tracking method using the color information of the image in surveillance network. This method perform a object detection using of adaptive color model. Object contour detection plays an important role in application such as object recognition. Experimental results demonstrate successful object detection over a wide range of object's variation in color and scale. In applications to detect an object in real time, when transmitting a large amount of image data it is possible to find the mode of a color distribution. The specific color of an object is modified at dynamically changing color in image. So, this algorithm detects the tracking area information of object within relevant tracking area and only tracking the movement of that object.Through experiments, we show that proposed method is more robust than other methods under certain ideal situations.

• Transactions on Semiconductor Engineering
• /
• v.2 no.1
• /
• pp.17-20
• /
• 2024

This paper presents a novel lightweight object detection model tailored for low-powered edge devices, addressing the limitations of traditional resource-intensive computer vision models. Our proposed detector, inspired by the Single Shot Detector (SSD), employs a compact yet robust network design. Crucially, it integrates an 'enhancer block' that significantly boosts its efficiency in detecting smaller objects. The model comprises two primary components: the Light_Block for efficient feature extraction using Depth-wise and Pointwise Convolution layers, and the Enhancer_Block for enhanced detection of tiny objects. Trained from scratch on the Udacity Annotated Dataset with image dimensions of 300x480, our model eschews the need for pre-trained classification weights. Weighing only 5.5MB with approximately 0.43M parameters, our detector achieved a mean average precision (mAP) of 27.7% and processed at 140 FPS, outperforming conventional models in both precision and efficiency. This research underscores the potential of lightweight designs in advancing object detection for edge devices without compromising accuracy.

• The KIPS Transactions:PartB
• /
• v.13B no.4 s.107
• /
• pp.361-368
• /
• 2006

In this paper we present a snake-based scheme for efficiently detecting contours of objects with boundary concavities. The proposed method is composed of two steps. First, the object's boundary is detected using the proposed snake model. Second, snake points are optimized by inserting new points and deleting unnecessary points to better describe the object's boundary. The proposed algorithm can successfully extract objects with boundary concavities. Experimental results have shown that our algorithm produces more accurate segmentation results than the conventional algorithm.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• fall
• /
• pp.113-116
• /
• 2021

일반적으로 딥러닝 기반의 객체 검출(Object Detection)기법은 합성곱 신경망(Convolutional Neural Network, CNN)을 통해 입력된 영상의 특징(Feature)을 추출하여 이를 통해 객체 검출을 수행한다. 최근 자연어 처리 분야에서 획기적인 성능을 보인 트랜스포머(Transformer)가 영상 분류, 객체 검출과 같은 컴퓨터 비전 작업을 수행하는데 있어 경쟁력이 있음이 드러나고 있다. 본 논문에서는 YOLOv4-CSP의 CSP 블록을 개선한 one-stage 방식의 객체 검출 네트워크를 제안한다. 개선된 CSP 블록은 트랜스포머(Transformer)의 멀티 헤드 어텐션(Multi-Head Attention)과 CSP 형태의 공간 피라미드 풀링(Spatial Pyramid Pooling, SPP) 연산을 기반으로 네트워크의 Backbone과 Neck에서의 feature 학습을 돕는다. 본 실험은 MSCOCO test-dev2017 데이터 셋으로 평가하였으며 제안하는 네트워크는 YOLOv4-CSP의 경량화 모델인 YOLOv4s-mish에 대하여 평균 정밀도(Average Precision, AP)기준 2.7% 향상된 검출 정확도를 보인다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.07a
• /
• pp.667-670
• /
• 2020

최근에는 증강현실, 로봇공학 등의 분야에서 객체의 위치 검출 이외에도, 객체의 자세에 대한 추정이 요구되고 있다. 객체의 자세 정보가 포함된 데이터셋은 위치 정보만 포함된 데이터셋에 비하여 상대적으로 매우 적기 때문에 인공 신경망 구조를 활용하기 어려운 측면이 있으나, 최근에 들어서는 기계학습 기반의 자세 추정 알고리즘들이 여럿 등장하고 있다. 본 논문에서는 이 가운데 Dense 6d Pose Object detector (DPOD) [11]의 구조를 기반으로 하여 가구의 조립 설명서에 그려진 가구 부품들의 자세를 추정하고자 한다. DPOD [11]는 입력으로 RGB 영상을 받으며, 해당 영상에서 자세를 추정하고자 하는 객체의 영역에 해당하는 픽셀들을 추정하고, 객체의 영역에 해당되는 각 픽셀에서 해당 객체의 3D 모델의 UV map 값을 추정한다. 이렇게 픽셀 개수만큼의 2D - 3D 대응이 생성된 이후에는, RANSAC과 PnP 알고리즘을 통해 RGB 영상에서의 객체와 객체의 3D 모델 간의 변환 관계 행렬이 구해지게 된다. 본 논문에서는 사전에 정해진 24개의 자세 후보들을 기반으로 가구 부품의 3D 모델을 2D에 투영한 RGB 영상들로 인공 신경망을 학습하였으며, 평가 시에는 실제 조립 설명서에서의 가구 부품의 자세를 추정하였다. 실험 결과 IKEA의 Stefan 의자 조립 설명서에 대하여 100%의 ADD score를 얻었으며, 추정 자세가 자세 후보군 중 정답 자세에 가장 근접한 경우를 정답으로 평가했을 때 100%의 정답률을 얻었다. 제안하는 신경망을 사용하였을 때, 가구 조립 설명서에서 가구 부품의 위치를 찾는 객체 검출기(object detection network)와, 각 개체의 종류를 구분하는 객체 리트리벌 네트워크(retrieval network)를 함께 사용하여 최종적으로 가구 부품의 자세를 추정할 수 있다.