• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.504-505
• /
• 2023

본 논문은 기존의 웨어러블 센서 방식이 아닌 영상으로 반려견의 행동을 분석하는 연구에 대한 것이다. 제안한 시스템은 영상에서 반려견의 영역을 탐지하고, 탐지된 이미지에서 반려견의 관절 좌표를 추출하여 행동을 판단하는 방식이다. 모든 프레임에 대해 처리하지 않고, 일정 주기 단위로 영상을 처리해 실시간성을 확보하였다. 제안한 시스템의 유용성은 실험으로 검증하였으며, 유의미한 실험 결과를 얻을 수 있었다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.1215-1217
• /
• 2023

가정에서 대형 폐기물을 배출하고 수거하는 과정에서 폐기물을 수동적으로 분류를 하는 것은 시간이 많이 소요되는 작업이다. 본 논문에서는 YOLOv4, 5, 7 모델을 비교하여 실생활에 사용가능한 대형 폐기물 탐지에 가장 적합한 모델을 찾는다. 이미지 증강 전 결과는 YOLOv7이 가장 좋은 성능을 보였다. 배출자가 촬영하는 각도나 위치, 시간 등의 변수를 고려하고자 증강을 시도하였고 증강 후 탐지 결과도 YOLOv7이 F1-score 93 %, mAP 96.6% 로 다른 모델보다 전체적으로 더 좋은 성능을 보였다.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.26 no.5
• /
• pp.392-399
• /
• 2023

This research paper investigates the effectiveness of using computer graphics(CG) based synthetic data for deep learning in military vehicle detection. In particular, we explore the use of synthetic image generation techniques to train deep neural networks for object detection tasks. Our approach involves the generation of a large dataset of synthetic images of military vehicles, which is then used to train a deep learning model. The resulting model is then evaluated on real-world images to measure its effectiveness. Our experimental results show that synthetic training data alone can achieve effective results in object detection. Our findings demonstrate the potential of CG-based synthetic data for deep learning and suggest its value as a tool for training models in a variety of applications, including military vehicle detection.

• The Journal of Bigdata
• /
• v.5 no.1
• /
• pp.29-39
• /
• 2020

Purpose The atmosphere of respect for individual tastes that have spread throughout society has changed the consumption trend. As a result, the travel industry is also seeing customized travel as a new trend that reflects consumers' personal tastes. In particular, there is a growing interest in 'film-induced tourism', one of the areas of travel industry. We hope to satisfy the individual's motivation for traveling while watching movies with customized travel proposals, which we expect to be a catalyst for the continued development of the 'film-induced tourism industry'. Design/methodology/approach In this study, we implemented a methodology through 'OCR' of extracting and suggesting film location information that viewers want to visit. First, we extract a scene from a movie selected by a user by using 'OpenCV', a real-time image processing library. In addition, we detected the location of characters in the scene image by using 'EAST model', a deep learning-based text area detection model. The detected images are preprocessed by using 'OpenCV built-in function' to increase recognition accuracy. Finally, after converting characters in images into recognizable text using 'Tesseract', an optical character recognition engine, the 'Google Map API' returns actual location information. Significance This research is significant in that it provides personalized tourism content using fourth industrial technology, in addition to existing film tourism. This could be used in the development of film-induced tourism packages with travel agencies in the future. It also implies the possibility of being used for inflow from abroad as well as to abroad.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.33 no.4
• /
• pp.225-235
• /
• 2020

Ultrasonic investigation of damage detection has been widely used for non-destructive testing of various concrete structures. This study focuses on damage detection analysis with the aid of wave propagation in two-phase composite concrete with aggregate (inclusion) and mortar (matrix). To fabricate a realistic simulation model containing a variety of irregular aggregate shapes, the mesh generation technique using an image processing technique was proposed. Initially, the domains and boundaries of the aggregates were extracted from the digital image of a typical concrete cut-section. This enables two different domains: aggregates and mortar in heterogeneous concrete sections, and applied the grids onto these domains to discretize the model. Subsequently, finite element meshes are generated in terms of spatial and temporal requirements of the model size. For improved analysis results, all meshes are designed to be quadrilateral type, and an additional process is conducted to improve the mesh quality. With this simulation model, wave propagation analyses were conducted with a central frequency of 75 kHz of the Mexican hat incident wave. Several void damages, such as needle-shaped cracks and void-shaped holes, were artificially introduced in the model. Finally, various formats of internal damage were detected by implementing energy mapping based signal processing.

• Journal of Broadcast Engineering
• /
• v.24 no.6
• /
• pp.983-991
• /
• 2019

This study aims to analyze movies which contain various stories according to the size of their shots. To achieve this, it is needed to classify dataset according to the shot size, such as extreme close-up shots, close-up shots, medium shots, full shots, and long shots. However, a typical video storytelling is mainly composed of close-up shots, medium shots, full shots, and long shots, it is not an easy task to construct an appropriate dataset for extreme close-up shots. To solve this, we propose an image cropping method based on the region of interest (ROI) detection. In this paper, we use the face detection and saliency detection to estimate the ROI. By cropping the ROI of close-up images, we generate extreme close-up images. The dataset which is enriched by proposed method is utilized to construct a model for classifying shots based on its size. The study can help to analyze the emotional changes of characters in video stories and to predict how the composition of the story changes over time. If AI is used more actively in the future in entertainment fields, it is expected to affect the automatic adjustment and creation of characters, dialogue, and image editing.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.4
• /
• pp.11-19
• /
• 2021

An abnormal object refers to a person, an object, or a mechanical device that performs abnormal and unusual behavior and needs observation or supervision. In order to detect this through artificial intelligence algorithm without continuous human intervention, a method of observing the specificity of temporal features using optical flow technique is widely used. In this study, an abnormal situation is identified by learning an algorithm that translates an input image frame to an optical flow image using a Generative Adversarial Network (GAN). In particular, we propose a technique that improves the pre-processing process to exclude unnecessary outliers and the post-processing process to increase the accuracy of identification in the test dataset after learning to improve the performance of the model's abnormal behavior identification. UCSD Pedestrian and UMN Unusual Crowd Activity were used as training datasets to detect abnormal behavior. For the proposed method, the frame-level AUC 0.9450 and EER 0.1317 were shown in the UCSD Ped2 dataset, which shows performance improvement compared to the models in the previous studies.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.47 no.3
• /
• pp.56-62
• /
• 2010

Weakly electric fish uses active sensing to detect the distortion of self-generated electric field in the underwater environments. The active electrolocation makes it possible to identify target objects from the surroundings without vision in the dark sea. Weakly electric fish have many electroreceptors over the whole body surface of electric fish, and sensor readings from a collection of electroreceptors are represented as an electric image. Many researchers have worked on finding features in the electric image to know how the weakly electric fish identify the target object. In this paper, we suggest a new mechanism of how the electrolocation can recognize a given target object among object plants. This approach is based on the differential components of the electric image, and has a potential to be applied to the underwater robotic system for object localization.

• Journal of Software Assessment and Valuation
• /
• v.16 no.2
• /
• pp.153-162
• /
• 2020

Digital image forgery detection is one of very important fields in the field of digital forensics. As the forged images change naturally through the advancement of technology, it has made it difficult to detect forged images. In this paper, we use passive forgery detection for copy paste forgery in digital images. In addition, it detects copy-paste forgery using the L₀ Norm-based LE operator, and compares the detection accuracy with the forgery detection using the existing L₂, L₁ Norm-based LE operator. In comparison of detection rates, the proposed lower triangular(Ayalneh and Choi) window was more robust to BAG mismatch detection than the conventional window filter. In addition, in the case of using the lower triangular window, the performance of image forgery detection was measured increasingly higher as the L₂, L₁ and L₀ Norm LE operator was performed.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.169-171
• /
• 2021

Image processing through cameras such as self-driving, CCTV, mobile phone security, and parking facilities is being used to solve many real-life problems. Simple classification is solved through image processing, but it is difficult to find images or in-image features of complexly mixed objects. To solve this feature point, we utilize deep learning techniques in classification, detection, and segmentation of image data so that we can think and judge closely. Of course, the results are better than just image processing, but we confirm that the results judged by the method of image segmentation using deep learning have deviations from the real object. In this paper, we study how to perform accuracy improvement through simple image processing just before outputting the output of deep learning image segmentation to increase the precision of image segmentation.