• 한국멀티미디어학회논문지
• /
• 제25권1호
• /
• pp.72-81
• /
• 2022

Visual Object Tracking is known as the most fundamental problem in the field of computer vision. Object tracking localize the region of target object with bounding box in the video. In this paper, a custom CNN is created to extract object feature that has strong and various information. This network was constructed as a Siamese network for use as a feature extractor. The input images are passed convolution block composed of a bottleneck layers, and features are emphasized. The feature map of the target object and the search area, extracted from the Siamese network, was input as a local proposal network. Estimate the object area using the feature map. The performance of the tracking algorithm was evaluated using the OTB2013 dataset. Success Plot and Precision Plot were used as evaluation matrix. As a result of the experiment, 0.611 in Success Plot and 0.831 in Precision Plot were achieved.

• 한국ITS학회 논문지
• /
• 제18권3호
• /
• pp.95-105
• /
• 2019

최근 들어 도로 노면 파손의 위치 정보를 수집하기 위한 영상 처리 기술에 대한 연구가 활발히 진행되고 있다. 대표적으로 차량에 탑재가 가능한 스마트폰이나 블랙박스를 통해 영상을 얻고 이를 영상처리 알고리즘을 사용하여 인식하는 기술이 주로 사용된다. GPS 모듈과 연계하여 실제 파손 위치를 파악할 때 가장 중요한 기술은 영상 처리 알고리즘인데, 근래에는 대부분 인공지능을 통한 알고리즘이 연구 주제로 주목받고 있다. 이와 같은 맥락에서 본 연구에서도 영역 기반의 합성곱 방식 계열의 객체인식 (Object Detection) 방법을 사용한 인공지능 영상 처리 알고리즘에 대하여 논의하고자 한다. 도로 노면 파손 객체 인식 성능을 향상시키기 위하여 도로 노면 파손 영상 600여 장과 일반적인 도로 주행 영상 1500여 장으로 학습 데이터베이스를 구성하였다. 또한 배경 객체 인식 방법을 적용한 지도 학습을 수행하여 도로 노면 파손의 오탐을 감소시켰다. 그 결과 동일한 테스트용 데이터베이스를 통해 알고리즘의 인식 성능을 mAP 평균값 기준 9.44%만큼 향상시킨 새로운 방법을 소개하고자 한다.

• Imaging Science in Dentistry
• /
• 제50권2호
• /
• pp.169-174
• /
• 2020

Purpose: Periodontal disease causes tooth loss and is associated with cardiovascular diseases, diabetes, and rheumatoid arthritis. The present study proposes using a deep learning-based object detection method to identify periodontally compromised teeth on digital panoramic radiographs. A faster regional convolutional neural network (faster R-CNN) which is a state-of-the-art deep detection network, was adapted from the natural image domain using a small annotated clinical data- set. Materials and Methods: In total, 100 digital panoramic radiographs of periodontally compromised patients were retrospectively collected from our hospital's information system and augmented. The periodontally compromised teeth found in each image were annotated by experts in periodontology to obtain the ground truth. The Keras library, which is written in Python, was used to train and test the model on a single NVidia 1080Ti GPU. The faster R-CNN model used a pretrained ResNet architecture. Results: The average precision rate of 0.81 demonstrated that there was a significant region of overlap between the predicted regions and the ground truth. The average recall rate of 0.80 showed that the periodontally compromised teeth regions generated by the detection method excluded healthiest teeth areas. In addition, the model achieved a sensitivity of 0.84, a specificity of 0.88 and an F-measure of 0.81. Conclusion: The faster R-CNN trained on a limited amount of labeled imaging data performed satisfactorily in detecting periodontally compromised teeth. The application of a faster R-CNN to assist in the detection of periodontally compromised teeth may reduce diagnostic effort by saving assessment time and allowing automated screening documentation.

• 방송공학회논문지
• /
• 제22권3호
• /
• pp.282-294
• /
• 2017

본 논문에서는 차량이 움직일 때 발생하는 카메라의 움직임, 도로상의 광원에 강건한 지능형 전조등 제어 시스템을 제안한다. 후보광원을 검출할 때 카메라의 원근 범위 추정 모델을 기반으로 한 ROI (Region of Interest)를 사용하며 이는 FROI (Front ROI)와 BROI (Back ROI)로 나뉘어 사용된다. ROI내에서 차량의 전조등과 후미등, 반사광 및 주변 도로의 조명들은 2개의 적응적 임계값에 의해 세그먼트화 된다. 세그먼트화 된 광원 후보군들로부터 후미등은 적색도(redness)와 Haar-like특징에 기반한 랜덤포레스트 분류기에 의해 검출된다. 전조등과 후미등 분류 과정에서 빠른 학습과 실시간 처리를 위해 SVM(Support Vector Machine) 또는 CNN(Convolutional Neural Network)을 사용하지 않고 랜덤포레스트 분류기를 사용했다. 마지막으로 페어링(Pairing) 단계에서는 수직좌표 유사성, 광원들간의 연관성 검사와 같은 사전 정의된 규칙을 적용한다. 제안된 알고리즘은 다양한 야간 운전환경을 포함하는 데이터에 적용한 결과, 최근의 관련연구 보다 향상된 검출 성능을 보여주었다.

• 방송공학회논문지
• /
• 제23권3호
• /
• pp.395-402
• /
• 2018

손바닥은 손금, 정맥 등 고유한 특징 정보를 포함하고 있는 신체 부위로 이를 이용한 다양한 사용자 인식 방법이 지속적으로 연구되어 왔다. 본 논문에서는 손금과 손바닥 정맥을 함께 이용한 사용자 인식 방법을 제안한다. 먼저, 손바닥 영역에서 손금과 정맥이 가장 많이 포함되어 있는 관심 영역을 검출하고, 에지 방향성 및 밝기 통계정보를 이용하여 정맥 영상 화질 개선을 수행한다. 이후 다중 스펙트럼 환경에서 획득된 복수의 영상을 각각 독립된 심층 신경망의 입력으로 이용하여 손금과 정맥 패턴을 효과적으로 학습한다. 다양한 상황에서의 실험을 통해 본 논문에서 제안하는 방법이 기존 사용자 인식 방법 대비 개선된 결과를 보임을 확인하고 그 결과를 분석한다.

• Smart Structures and Systems
• /
• 제30권4호
• /
• pp.385-395
• /
• 2022

Detecting cracks on a concrete structure is crucial for structural maintenance, a crack being an indicator of possible damage. Conventional crack detection methods which include visual inspection and non-destructive equipment, are typically limited to a small region and require time-consuming processes. Recently, to reduce the human intervention in the inspections, various researchers have sought computer vision-based crack analyses: One class is filter-based methods, which effectively transforms the image to detect crack edges. The other class is using deep-learning algorithms. For example, convolutional neural networks have shown high precision in identifying cracks in an image. However, when the objective is to classify not only the existence of crack but also the types of cracks, only a few studies have been reported, limiting their practical use. Thus, the presented study develops an image processing procedure that detects cracks and classifies crack types; whether the image contains a crazing-type, single crack, or multiple cracks. The properties and steps in the algorithm have been developed using field-obtained images. Subsequently, the algorithm is validated from additional 227 images obtained from an open database. For test datasets, the proposed algorithm showed accuracy of 92.8% in average. In summary, the developed algorithm can precisely classify crazing-type images, while some single crack images may misclassify into multiple cracks, yielding conservative results. As a result, the successful results of the presented study show potentials of using vision-based technologies for providing crack information with reduced human intervention.

• International Journal of Computer Science & Network Security
• /
• 제22권10호
• /
• pp.73-82
• /
• 2022

Early detection continues to be the mainstay of breast cancer control as well as the improvement of its treatment. Even so, the absence of cancer symptoms at the onset has early detection quite challenging. Therefore, various researchers continue to focus on cancer as a topic of health to try and make improvements from the perspectives of diagnosis, prevention, and treatment. This research's chief goal is development of a system with deep learning for classification of the breast cancer as non-malignant and malignant using mammogram images. The following two distinct approaches: the first one with the utilization of patches of the Region of Interest (ROI), and the second one with the utilization of the overall images is used. The proposed system is composed of the following two distinct stages: the pre-processing stage and the Convolution Neural Network (CNN) building stage. Of late, the use of meta-heuristic optimization algorithms has accomplished a lot of progress in resolving these problems. Teaching-Learning Based Optimization algorithm (TIBO) meta-heuristic was originally employed for resolving problems of continuous optimization. This work has offered the proposals of novel methods for training the Residual Network (ResNet) as well as the CNN based on the TLBO and the Genetic Algorithm (GA). The classification of breast cancer can be enhanced with direct application of the hybrid TLBO- GA. For this hybrid algorithm, the TLBO, i.e., a core component, will combine the following three distinct operators of the GA: coding, crossover, and mutation. In the TLBO, there is a representation of the optimization solutions as students. On the other hand, the hybrid TLBO-GA will have further division of the students as follows: the top students, the ordinary students, and the poor students. The experiments demonstrated that the proposed hybrid TLBO-GA is more effective than TLBO and GA.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2017년도 추계학술대회
• /
• pp.166-169
• /
• 2017

현재 딥러닝을 이용한 연구들이 활발하게 이뤄지고 있고, 많은 분야에서 좋은 결과를 보여주고 있다. PCB(Printed Circuit Board) 기판 위에 실장 된 부품을 인식할 때 템플릿 매칭을 이용한 방식이 주를 이룬다. 하지만 템플릿 매칭은 모양과 방향, 밝기에 따라 여러 템플릿이 존재해야하고, 영상 전체를 탐색하여 매칭하기 때문에 수행시간이 오래 걸린다. 또한 인식률이 상당히 떨어지는 단점이 존재한다. 이로 인해 본 논문에서는 하나의 영상에서 여러 개의 물체를 분류할 때 사용하는 기계학습 방법 중 하나인 Faster-RCNN(Region-based Convolutional Neural Networks)을 이용하여 PCB 부품들을 인식하는 방식을 사용하였으며, 이 방법은 템플릿 매칭 방식보다 수행시간과 인식 면에서 더욱 좋은 성능을 보여준다.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2023년도 학술발표회
• /
• pp.173-173
• /
• 2023

The Mekong River Basin (MRB) is a crucial watershed in Asia, impacting over 60 million people across six developing nations. Accurate satellite-based precipitation products (SPPs) are essential for effective hydrological and watershed management in this region. However, the performance of SPPs has been varied and limited. The APHRODITE product, a unique gauge-based dataset for MRB, is widely used but is only available until 2015. In this study, we present a novel framework for correcting SPPs in the MRB by employing a deep learning approach that combines convolutional neural networks and encoder-decoder architecture to address pixel-by-pixel bias and enhance accuracy. The DLF was applied to four widely used SPPs (TRMM, CMORPH, CHIRPS, and PERSIANN-CDR) in MRB. For the original SPPs, the TRMM product outperformed the other SPPs. Results revealed that the DLF effectively bridged the spatial-temporal gap between the SPPs and the gauge-based dataset (APHRODITE). Among the four corrected products, ADJ-TRMM demonstrated the best performance, followed by ADJ-CDR, ADJ-CHIRPS, and ADJ-CMORPH. The DLF offered a robust and adaptable solution for bias correction in the MRB and beyond, capable of detecting intricate patterns and learning from data to make appropriate adjustments. With the discontinuation of the APHRODITE product, DLF represents a promising solution for generating a more current and reliable dataset for MRB research. This research showcased the potential of deep learning-based methods for improving the accuracy of SPPs, particularly in regions like the MRB, where gauge-based datasets are limited or discontinued.

• 국제학술발표논문집
• /
• The 9th International Conference on Construction Engineering and Project Management
• /
• pp.877-885
• /
• 2022

Vision-based activity recognition has been widely attempted at construction sites to estimate productivity and enhance workers' health and safety. Previous studies have focused on extracting an individual worker's postural information from sequential image frames for activity recognition. However, various trades of workers perform different tasks with similar postural patterns, which degrades the performance of activity recognition based on postural information. To this end, this research exploited a concept of human-object interaction, the interaction between a worker and their surrounding objects, considering the fact that trade workers interact with a specific object (e.g., working tools or construction materials) relevant to their trades. This research developed an approach to understand the context from sequential image frames based on four features: posture, object, spatial features, and temporal feature. Both posture and object features were used to analyze the interaction between the worker and the target object, and the other two features were used to detect movements from the entire region of image frames in both temporal and spatial domains. The developed approach used convolutional neural networks (CNN) for feature extractors and activity classifiers and long short-term memory (LSTM) was also used as an activity classifier. The developed approach provided an average accuracy of 85.96% for classifying 12 target construction tasks performed by two trades of workers, which was higher than two benchmark models. This experimental result indicated that integrating a concept of the human-object interaction offers great benefits in activity recognition when various trade workers coexist in a scene.