• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.16 no.6
• /
• pp.392-399
• /
• 2023

This paper proposes a deep learning model to determine the region and depth of cabbage cores for robotic automation of the cabbage core removal process during the kimchi manufacturing process. In addition, rather than predicting the depth of the measured cabbage, a model was presented that simultaneously detects and classifies the area by converting it into a discrete class. For deep learning model learning and verification, RGB images of the harvested cabbage 522 were obtained. The core region and depth labeling and data augmentation techniques from the acquired images was processed. MAP, IoU, acuity, sensitivity, specificity, and F1-score were selected to evaluate the performance of the proposed YOLO-v4 deep learning model-based cabbage core area detection and classification model. As a result, the mAP and IoU values were 0.97 and 0.91, respectively, and the acuity and F1-score values were 96.2% and 95.5% for depth classification, respectively. Through the results of this study, it was confirmed that the depth information of cabbage can be classified, and that it can be used in the development of a robot-automation system for the cabbage core removal process in the future.

• Korean Journal of Remote Sensing
• /
• v.39 no.3
• /
• pp.297-309
• /
• 2023

The degradation of coastal ecosystems and fishery environments is accelerating due to the recent phenomenon of invertebrate grazers. To effectively monitor and implement preventive measures for this phenomenon, the adoption of remote sensing-based monitoring technology for extensive maritime areas is imperative. In this study, we compared and analyzed the robustness of deep learning-based object detection modelsfor detecting and monitoring invertebrate grazersfrom underwater videos. We constructed an image dataset targeting seven representative species of invertebrate grazers in the coastal waters of South Korea and trained deep learning-based object detection models, You Only Look Once (YOLO)v7 and YOLOv8, using this dataset. We evaluated the detection performance and speed of a total of six YOLO models (YOLOv7, YOLOv7x, YOLOv8s, YOLOv8m, YOLOv8l, YOLOv8x) and conducted robustness evaluations considering various image distortions that may occur during underwater filming. The evaluation results showed that the YOLOv8 models demonstrated higher detection speed (approximately 71 to 141 FPS [frame per second]) compared to the number of parameters. In terms of detection performance, the YOLOv8 models (mean average precision [mAP] 0.848 to 0.882) exhibited better performance than the YOLOv7 models (mAP 0.847 to 0.850). Regarding model robustness, it was observed that the YOLOv7 models were more robust to shape distortions, while the YOLOv8 models were relatively more robust to color distortions. Therefore, considering that shape distortions occur less frequently in underwater video recordings while color distortions are more frequent in coastal areas, it can be concluded that utilizing YOLOv8 models is a valid choice for invertebrate grazer detection and monitoring in coastal waters.

• Journal of Broadcast Engineering
• /
• v.26 no.4
• /
• pp.453-462
• /
• 2021

A lot of license plate data is required for car number recognition. License plate data needs to be balanced from past license plates to the latest license plates. However, it is difficult to obtain data from the actual past license plate to the latest ones. In order to solve this problem, a license plate recognition study through deep learning is being conducted by creating a synthetic license plates. Since the synthetic data have differences from real data, and various data augmentation techniques are used to solve these problems. Existing data augmentation simply used methods such as brightness, rotation, affine transformation, blur, and noise. In this paper, we apply a style transformation method that transforms synthetic data into real-world data styles with data augmentation methods. In addition, real license plate data are noisy when it is captured from a distance and under the dark environment. If we simply recognize characters with input data, chances of misrecognition are high. To improve character recognition, in this paper, we applied the DeblurGANv2 method as a quality improvement method for character recognition, increasing the accuracy of license plate recognition. The method of deep learning for license plate detection and license plate number recognition used YOLO-V5. To determine the performance of the synthetic license plate data, we construct a test set by collecting our own secured license plates. License plate detection without style conversion recorded 0.614 mAP. As a result of applying the style transformation, we confirm that the license plate detection performance was improved by recording 0.679mAP. In addition, the successul detection rate without image enhancement was 0.872, and the detection rate was 0.915 after image enhancement, confirming that the performance improved.

• International journal of advanced smart convergence
• /
• v.10 no.4
• /
• pp.110-116
• /
• 2021

This paper proposed a real-time earlobe detection system using deep learning on the web. Existing deep learning-based detection methods often find independent objects such as cars, mugs, cats, and people. We proposed a way to receive an image through the camera of the user device in a web environment and detect the earlobe on the server. First, we took a picture of the user's face with the user's device camera on the web so that the user's ears were visible. After that, we sent the photographed user's face to the server to find the earlobe. Based on the detected results, we printed an earring model on the user's earlobe on the web. We trained an existing YOLO v5 model using a dataset of about 200 that created a bounding box on the earlobe. We estimated the position of the earlobe through a trained deep learning model. Through this process, we proposed a real-time earlobe detection system on the web. The proposed method showed the performance of detecting earlobes in real-time and loading 3D models from the web in real-time.

• Korean Journal of Remote Sensing
• /
• v.40 no.1
• /
• pp.33-43
• /
• 2024

This research assessed the applicability of the You Only Look Once (YOLO)v8 and DeepLabv3+ models for the effective detection of compost heaps, identified as a significant source of non-point source pollution. Utilizing high-resolution imagery acquired through Unmanned Aerial Vehicles(UAVs), the study conducted a comprehensive comparison and analysis of the quantitative and qualitative performances. In the quantitative evaluation, the YOLOv8 model demonstrated superior performance across various metrics, particularly in its ability to accurately distinguish the presence or absence of covers on compost heaps. These outcomes imply that the YOLOv8 model is highly effective in the precise detection and classification of compost heaps, thereby providing a novel approach for assessing the management grades of compost heaps and contributing to non-point source pollution management. This study suggests that utilizing UAVs and deep learning technologies for detecting and managing compost heaps can address the constraints linked to traditional field survey methods, thereby facilitating the establishment of accurate and effective non-point source pollution management strategies, and contributing to the safeguarding of aquatic environments.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.4
• /
• pp.49-57
• /
• 2022

The government is making great efforts to prevent crimes such as missing children by using public CCTVs. However, there is a shortage of operating manpower, weakening of concentration due to long-term concentration, and difficulty in tracking. In addition, applying real-time object search, re-identification, and tracking through a deep learning algorithm showed a phenomenon of increased parameters and insufficient memory for speed reduction due to complex network analysis. In this paper, we designed the network to improve speed and save memory through the application of Yolo v4, which can recognize real-time objects, and the application of Batch and TensorRT technology. In this thesis, based on the research on these advanced algorithms, OSNet re-ranking and K-reciprocal nearest neighbor for re-identification, Jaccard distance dissimilarity measurement algorithm for correlation, etc. are developed and used in the solution of CCTV national safety identification and tracking system. As a result, we propose a solution that can track objects by recognizing and re-identification objects in real-time within situation of a Korean public multi-CCTV environment through a set of algorithm combinations.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2022.07a
• /
• pp.45-48
• /
• 2022

건축물 안전 점검은 대부분 전문가의 현장 방문을 통한 육안검사다. 그중 균열 검사는 건물 위험도를 나타내는 중요한 지표로써 발생 위치, 진행성, 크기를 조사하는데, 최근 균열 조사 방식에 대해 객관성과 체계성을 보완할 딥러닝 개발이 활발하다. 그러나 균열 이미지는 외부 현장에 모양, 규모도 많은 종류라 도메인이 다양해야 하는데 대부분 제한된 환경과 실제적인 균열 검사와는 무관한 데이터로 구성되어 실효적이지 않다. 본 연구에서는 균열 조사에 적합하고 Wild 환경에 적용 가능한 POC 데이터셋을 소개한다. 기존 균열 공인 데이터셋 4종의 특징과 한계점을 분석을 토대로 고해상도 이미지로써 균열의 세부 특징을 담았고 균열 유사 환경과 조건들을 추가 촬영해 균열 검출에 강인하게 학습되도록 지향하였다. 정제 및 라벨링 작업을 거친 POC 데이터 셋은 균열 검출모델인 YOLO-v5으로 성능을 실험하였고, mAP(mean Average Precision) 75.5%로 높은 검출률을 보였다. POC 데이터셋으로 더욱 도메인에 적응적(Domain-adapted)인 인공지능 모델을 개발하여 건물, 댐, 교량 등 각종 대형 건축물에 대한 안전하고 효과적인 안전 관리 도구로써 활용할 것을 기대한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.957-958
• /
• 2023

본 연구에서는 공정 내 인체 끼임 사고 및 저출산 고령화 시대 등의 문제들로 인한 인력 감소 문제들을 해결하기 위하여 인공지능 기반의 모빌리티 플랫폼을 개발하였다. 본 플랫폼은 yolo-v4 기반으로 로봇이 공정 내부를 이동하면서 공정 내 불량 와이어를 검출하여 공정 유지보수 관리자에게 알려 주고 실시간으로 공정 내부 상황을 무선으로 모니터링할 수 있는 기술을 연구하였다. 또한, 로봇에는 무선 충전 기능을 갖추고 있어 추가적인 공정 내 인체 끼임 사고 등을 예방할 수 있도록 하였다. 본 연구 결과에서는 불량 와이어 검출 시, 평균 92.1%의 정확도를 보였으며, 공정 제품에 대한 양품, 불량품의 검출은 평균 98%의 정확도를 보였다. 또한, 실시간 무선 영상 모니터링는 24fps로 전송되어 공정 내부를 살피는 것에 유의미한 결과를 보였다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.137-137
• /
• 2023

해양쓰레기는 해안의 심미적 가치 저하뿐만 아니라 생태계 파괴, 유령 어업에 따른 수산업 피해 등의 사회적·환경적 문제를 발생시키며, 그중 70% 이상은 육상 기인으로 플라스틱 및 기타 쓰레기가 주를 이루는 해외와 달리 국내의 경우 다량의 초목류를 포함하고 있다. 다양한 부유쓰레기에 대한 기존의 해양쓰레기량 추정의 한계와 하천·하구 쓰레기 수거의 효율화를 위해 해양으로 유입되는 부유쓰레기 방지를 위한 실효성 있는 대책 수립이 필요한 실정이다. 본 연구는 해양 유입 전 하천의 차단시설에 차집된 부유쓰레기의 수거 효율화 및 지속가능한 해양쓰레기 데이터 구축을 위해 AI기반의 기술을 통해 부유쓰레기 성상 분석 기법(Object Detection)과 차집량 분석 기법(Semantic Segmentation)을 활용하였다. 실제와 유사한 데이터 수집을 위해 다양한 하천 환경(정수조, 소하천, 급경사수로)에 대해 탁도(녹조, 유사), 광량, 쓰레기형상, 초목류 함량, 날씨(소하천), 유속(급경사수로) 등의 실험조건에 대하여 해양쓰레기 분류 기준 및 통계를 바탕으로 부유쓰레기 종류 선정하여 학습을 위한 데이터를 수집하였다. 학습 목적에 따라 구분하여 라벨링(Bounding box, Polygon)을 수행하고, 각 분석 기법별 전이학습을 통해 Phase 1(정수조), Phase 2(소하천), Phase 3(급경사수로) 순서로 모델을 고도화하였다. 성상 분석을 위해 YOLO v4를 활용하여 Train, Test DataSet(9:1)을 구성하고 학습 및 평가는 Iteration마다의 mAP, loss 값을 통해 비교하였으며, 학습 Phase에 따라 모델 고도화로 Test Set의 mAP 값이 성상별로 높아짐을 확인하였으며, 차집량 분석을 위해 Unet을 활용하여 Train, Test, Validation DataSet(8.5:1:0.5)을 구성하고 epoch별 IoU(intersection over Union), F1-score, loss 값을 비교하여 정성적, 정량적 평가 모두 Phase 3에서 가장 높은 성능을 확인하였다. 향후 하천 환경에서의 다양한 영양인자별 분석을 통해 주요 영향인자 도출 및 Hyper Parameter 최적화를 통한 모델 고도화로 인해 활용성이 높아질 것으로 판단된다.

• Korean Journal of Remote Sensing
• /
• v.40 no.1
• /
• pp.115-122
• /
• 2024

Multi-object tracking (MOT) is a vital component in understanding the surrounding environments. Previous research has demonstrated that MOT can successfully detect and track surrounding objects. Nonetheless, inaccurate classification of the tracking objects remains a challenge that needs to be solved. When an object approaching from a distance is recognized, not only detection and tracking but also classification to determine the level of risk must be performed. However, considering the erroneous classification results obtained from the detection as the track class can lead to performance degradation problems. In this paper, we discuss the limitations of classification in tracking under the classification uncertainty of the detector. To address this problem, a class update module is proposed, which leverages the class uncertainty estimation of the detector to mitigate the classification error of the tracker. We evaluated our approach on the VisDrone-MOT2021 dataset,which includes multi-class and uncertain far-distance object tracking. We show that our method has low certainty at a distant object, and quickly classifies the class as the object approaches and the level of certainty increases.In this manner, our method outperforms previous approaches across different detectors. In particular, the You Only Look Once (YOLO)v8 detector shows a notable enhancement of 4.33 multi-object tracking accuracy (MOTA) in comparison to the previous state-of-the-art method. This intuitive insight improves MOT to track approaching objects from a distance and quickly classify them.