• International journal of advanced smart convergence
• /
• 제11권2호
• /
• pp.22-29
• /
• 2022

In this paper, for the purpose of designing an real-time unmanned monitoring system, the YOLOv5s (small) object detection model was applied on the NVIDIA TX2^TM AI (Artificial Intelligence) edge computing platform in order to design the fundamental function of an unmanned monitoring system that can detect objects in real time. YOLOv5s was applied to the our real-time unmanned monitoring system based on the performance evaluation of object detection algorithms (for example, R-CNN, SSD, RetinaNet, and YOLOv5). In addition, the performance of the four YOLOv5 models (small, medium, large, and xlarge) was compared and evaluated. Furthermore, based on these results, the YOLOv5s model suitable for the design purpose of this paper was ported to the NVIDIA TX2^TM AI edge computing system and it was confirmed that it operates normally. The real-time unmanned monitoring system designed as a result of the research can be applied to various application fields such as an security or monitoring system. Future research is to apply NMS (Non-Maximum Suppression) modification, model reconstruction, and parallel processing programming techniques using CUDA (Compute Unified Device Architecture) for the improvement of object detection speed and performance.

• 국제학술발표논문집
• /
• The 9th International Conference on Construction Engineering and Project Management
• /
• pp.893-903
• /
• 2022

Recently, computer vision (CV)-based safety monitoring (i.e., object detection) system has been widely researched in the construction industry. Sufficient and high-quality data collection is required to detect objects accurately. Such data collection is significant for detecting small objects or images from different camera angles. Although several previous studies proposed novel data augmentation and synthetic data generation approaches, it is still not thoroughly addressed (i.e., limited accuracy) in the dynamic construction work environment. In this study, we proposed a game engine-driven synthetic data generation model to enhance the accuracy of the CV-based object detection model, mainly targeting small objects. In the virtual 3D environment, we generated synthetic data to complement training images by altering the virtual camera angles. The main contribution of this paper is to confirm whether synthetic data generated in the game engine can improve the accuracy of the CV-based object detection model.

• Journal of Information Processing Systems
• /
• 제9권4호
• /
• pp.592-601
• /
• 2013

Object segmentation is a challenging task in image processing and computer vision. In this paper, we present a visual attention based segmentation method to segment small sized interesting objects in natural images. Different from the traditional methods, we first search the region of interest by using our novel saliency-based method, which is mainly based on band-pass filtering, to obtain the appropriate frequency. Secondly, we applied the Gaussian Mixture Model (GMM) to locate the object region. By incorporating the visual attention analysis into object segmentation, our proposed approach is able to narrow the search region for object segmentation, so that the accuracy is increased and the computational complexity is reduced. The experimental results indicate that our proposed approach is efficient for object segmentation in natural images, especially for small objects. Our proposed method significantly outperforms traditional GMM based segmentation.

• 한국해양공학회지
• /
• 제35권1호
• /
• pp.91-97
• /
• 2021

Marine accidents caused by ships have brought about economic and social losses as well as human casualties. Most of these accidents are caused by small and medium-sized ships and are due to their poor conditions and insufficient equipment compared with larger vessels. Measures are quickly needed to improve the conditions. This paper discusses a video-integrated collision prediction and fall detection system to support the safe navigation of small- and medium-sized ships. The system predicts the collision of ships and detects falls by crew members using the CCTV, displays the analyzed integrated information using automatic identification system (AIS) messages, and provides alerts for the risks identified. The design consists of an object recognition algorithm, interface module, integrated display module, collision prediction and fall detection module, and an alarm management module. For the basic research, we implemented a deep learning algorithm to recognize the ship and crew from images, and an interface module to manage messages from AIS. To verify the implemented algorithm, we conducted tests using 120 images. Object recognition performance is calculated as mAP by comparing the pre-defined object with the object recognized through the algorithms. As results, the object recognition performance of the ship and the crew were approximately 50.44 mAP and 46.76 mAP each. The interface module showed that messages from the installed AIS were accurately converted according to the international standard. Therefore, we implemented an object recognition algorithm and interface module in the designed collision prediction and fall detection system and validated their usability with testing.

• 한국컴퓨터정보학회논문지
• /
• 제25권1호
• /
• pp.45-53
• /
• 2020

기존의 CNN 알고리즘은 위성영상과 같은 대형 이미지에서 소형 객체를 식별하는 것이 불가능하다는 문제점을 가지고 있었다. 본 연구에서는 이러한 문제를 해결하기 위해 관심영역 설정 및 이미지 분할 기법을 적용한 CNN 알고리즘 개선방안을 제시하였다. 실험은 비행장 및 항공기 데이터셋으로 전환학습한 YOLOv3 / Faster R-CNN 알고리즘과 테스트용 대형 이미지를 이용하여 진행하였으며, 우선 대형 이미지에서 관심영역을 식별하고 이를 순차적으로 분할해 나가며 CNN 알고리즘의 객체식별 결과를 비교하였다. 분할 이미지의 크기는 실험을 통해 최소 분할로 최대의 식별률을 얻을 수 있는 최적의 이미지 조각 크기를 도출하여 적용하였다. 실험 결과, 본 연구에서 제시한 방안을 통해 CNN 알고리즘으로 대형 이미지에서의 소형 객체를 식별하는 것이 충분히 가능함을 검증하였다.

• 한국인터넷방송통신학회논문지
• /
• 제21권3호
• /
• pp.81-89
• /
• 2021

자연 조건에서 Apple 감지에는 가림 문제와 작은 대상 감지 어려움이 있다. 본 논문은 SSD 기반의 개선 된 모델을 제안한다. SSD 백본 네트워크 VGG16은 ResNet50 네트워크 모델로 대체되고 수용 필드 구조 RFB 구조가 도입되었다. RFB 모델은 작은 표적의 특징 정보를 증폭하고 작은 표적의 탐지 정확도를 향상시킨다. 유지해야 하는 정보를 필터링하기 위해 주의 메커니즘 (SE)과 결합하면 감지 대상의 의미 정보가 향상된다. 향상된 SSD 알고리즘은 VOC2007 데이터 세트에 대해 학습된다. SSD에 비해 개선 된 알고리즘은 폐색 및 작은 표적 탐지의 정확도를 3.4 % 및 3.9 % 향상 시켰다. 이 알고리즘은 오 탐지율과 누락된 감지율을 향상 시켰다. 본 논문에서 제안한 개선 된 알고리즘은 더 높은 효율성을 갖는다.

• 로봇학회논문지
• /
• 제14권1호
• /
• pp.1-7
• /
• 2019

Recently, smart factories have attracted much attention as a result of the 4th Industrial Revolution. Existing factory automation technologies are generally designed for simple repetition without using vision sensors. Even small object assemblies are still dependent on manual work. To satisfy the needs for replacing the existing system with new technology such as bin picking and visual servoing, precision and real-time application should be core. Therefore in our work we focused on the core elements by using deep learning algorithm to detect and classify the target object for real-time and analyzing the object features. We chose YOLO CNN which is capable of real-time working and combining the two tasks as mentioned above though there are lots of good deep learning algorithms such as Mask R-CNN and Fast R-CNN. Then through the line and inside features extracted from target object, we can obtain final outline and estimate object posture.

• 대한임베디드공학회논문지
• /
• 제18권5호
• /
• pp.203-208
• /
• 2023

Recently, with the growing interest in the field of autonomous driving, many researchers have been focusing on developing autonomous driving software platforms. In particular, we have concentrated on developing 3D object detection models that can improve real-time performance. In this paper, we introduce a self-constructed 3D LiDAR dataset specific to domestic environments and propose a VariFocal-based CenterPoint for the 3D object detection model, with improved performance over the previous models. Furthermore, we present experimental results comparing the performance of the 3D object detection modules using our self-built and public dataset. As the results show, our model, which was trained on a large amount of self-constructed dataset, successfully solves the issue of failing to detect large vehicles and small objects such as motorcycles and pedestrians, which the previous models had difficulty detecting. Consequently, the proposed model shows a performance improvement of about 1.0 mAP over the previous model.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권9호
• /
• pp.4386-4404
• /
• 2016

Most existing salient object detection algorithms commonly employed segmentation techniques to eliminate background noise and reduce computation by treating each segment as a processing unit. However, individual small segments provide little information about global contents. Such schemes have limited capability on modeling global perceptual phenomena. In this paper, a novel salient object detection algorithm is proposed based on region merging. An adaptive-based merging scheme is developed to reassemble regions based on their color dissimilarities. The merging strategy can be described as that a region R is merged with its adjacent region Q if Q has the lowest dissimilarity with Q among all Q's adjacent regions. To guide the merging process, superpixels that located at the boundary of the image are treated as the seeds. However, it is possible for a boundary in the input image to be occupied by the foreground object. To avoid this case, we optimize the boundary influences by locating and eliminating erroneous boundaries before the region merging. We show that even though three simple region saliency measurements are adopted for each region, encouraging performance can be obtained. Experiments on four benchmark datasets including MSRA-B, SOD, SED and iCoSeg show the proposed method results in uniform object enhancement and achieve state-of-the-art performance by comparing with nine existing methods.

• 한국정보컨버전스학회:학술대회논문집
• /
• 한국정보컨버전스학회 2008년도 International conference on information convergence
• /
• pp.103-110
• /
• 2008

In this thesis, we describe a statistical method for 3D object detection. In this method, we decompose the 3D geometry of each object into a small number of viewpoints. For each viewpoint, we construct a decision rule that determines if the object is present at that specific orientation. Each decision rule uses the statistics of both object appearance and "non-object" visual appearance. We represent each set of statistics using a product of histograms. Each histogram represents the joint statistics of a subset of wavelet coefficients and their position on the object. Our approach is to use many such histograms representing a wide variety of visual attributes. Using this method, we have developed the first algorithm that can reliably detect faces that vary from frontal view to full profile view and the first algorithm that can reliably detect cars over a wide range of viewpoints.