• Geophysics and Geophysical Exploration
• /
• v.26 no.4
• /
• pp.211-228
• /
• 2023

Ground-penetrating radar (GPR) surveys are commonly used to monitor embankments, which is a nondestructive geophysical method. The results of GPR surveys can be complex, depending on the situation, and data processing and interpretation are subject to expert experiences, potentially resulting in false detection. Additionally, this process is time-intensive. Consequently, various studies have been undertaken to detect cavities in GPR survey data using deep learning methods. Deep-learning-based approaches require abundant data for training, but GPR field survey data are often scarce due to cost and other factors constaining field studies. Therefore, in this study, a deep- learning-based model was developed for embankment GPR survey cavity detection using data augmentation strategies. A dataset was constructed by collecting survey data over several years from the same embankment. A you look only once (YOLO) model, commonly used in computer vision for object detection, was employed for this purpose. By comparing and analyzing various strategies, the optimal data augmentation approach was determined. After initial model development, a stepwise process was employed, including box clustering, transfer learning, self-ensemble, and model ensemble techniques, to enhance the final model performance. The model performance was evaluated, with the results demonstrating its effectiveness in detecting cavities in embankment GPR survey data.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.32 no.5
• /
• pp.495-503
• /
• 2014

As the number of vehicles has been sharply increases, the significance of safety and effective operation issues in the parking lot is being emphasized, which takes a part of the transportation system. Recently, there have been several studies for the parking management by detecting moving object, however, recognizing numbers of fast-moving vehicles simultaneously in the picture is still a challenging problem. The parking lot in public area, or large-sized buildings has clear parking section, whereas the sensor system is configured to monitor a plurality of parking spaces. Therefore, by considering those parking lots, we suggested to develop the real-time parking availability information system by applying the real-time image processing techniques. with the help of template matching. Following the study, we wanted to provide the alternative method for parking management system through the reference template makers by recognizing movements of parked vehicles with the size and shape, regardless of direct detecting of driving movements. In addition, we evaluated the applicability and performances of the information system, presented in this study, and implemented a prototype system to simulate the parking statuses of each floor. In fat, it was possible to manage and analyze statistics about the total number of parking spaces and the number of vehicles parked through real-time video flames. We expected that the result of the study will be advanced, following the user-friendliness and cost reduction in operating parking management system and giving information by efficient analysis of parking situation.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.4
• /
• pp.101-109
• /
• 2014

It is an essential element for the establishment of image processing related systems to find the exact contour from the image of an arbitrary object. In particular, if a vision system is established to inspect the products in the automated production process, it is very important to detect the contours for standardized shapes such lines and curves. In this paper, we propose a prototype adaptive dynamic contour model, p-Snake with improved contour extraction algorithms by adding the prototype energy. The proposed method is to find the initial contour by applying the existing Snake algorithm after Sobel operation is performed for prototype analysis. Next, the final contour of the object is detected by analyzing prototypes such as lines and circles, defining prototype energy and using it as an additional energy item in the existing Snake function on the basis of information on initial contour. We performed experiments on 340 images obtained by using an environment that duplicated the background of an industrial site. It was found that even if objects are not clearly distinguished from the background due to noise and lighting or the edges being insufficiently visible in the images, the contour can be extracted. In addition, in the case of similarity which is the measure representing how much it matches the prototype, the prototype similarity of contour extracted from the proposed p-ACM is superior to that of ACM by 9.85%.

• The Journal of the Korea Contents Association
• /
• v.16 no.7
• /
• pp.689-699
• /
• 2016

As the demand for LCD increases, the importance of inspection equipment for improving the efficiency of LCD production is continuously emphasized. The pattern inspection apparatus is one that detects minute defects of pattern quickly using optical equipment such as line scan camera. This pattern inspection apparatus makes a decision on whether a pixel is a defect or not using a single threshold value in order to meet constraint of real time inspection. However, a method that uses an adaptive thresholding scheme with different threshold values according to characteristics of each region in a pattern can greatly improve the performance of defect detection. To apply this adaptive thresholding scheme it has to be known that a certain pixel to be inspected belongs to which region. Therefore, this paper proposes a region matching algorithm that recognizes the region of each pixel to be inspected. The proposed algorithm is based on the pattern matching scheme with the consideration of real time constraint of machine vision and implemented through GPGPU in order to be applied to a practical system. Simulation results show that the proposed method not only satisfies the requirement for processing time of practical system but also improves the performance of defect detection.

• KIPS Transactions on Software and Data Engineering
• /
• v.4 no.10
• /
• pp.465-472
• /
• 2015

We propose a novel plane detection in disparity space and evaluate its performance. Our method simplifies and makes scenes in disparity space easily dealt with by approximating various surfaces as planes. Moreover, the approximated planes can be represented in the same size as in the real world, and can be employed for obstacle detection and camera pose estimation. Using a stereo matching technique, our method first creates a disparity image which consists of binocular disparity values at xy-coordinates in the image. Slants of disparity values are estimated by exploiting a line simplification algorithm which allows our method to reflect global changes against x or y axis. According to pairs of x and y slants, we label the disparity image. 4-connected disparities with the same label are grouped, on which least squared model estimates plane parameters. N plane models with the largest group of disparity values which satisfy their plane parameters are chosen. We quantitatively and qualitatively evaluate our plane detection. The result shows 97.9%와 86.6% of quality in our experiment respectively on cones and cylinders. Proposed method excellently extracts planes from Middlebury and KITTI dataset which are typically used for evaluation of stereo matching algorithms.

• Journal of the Institute of Convergence Signal Processing
• /
• v.22 no.4
• /
• pp.179-184
• /
• 2021

Along with the development of the fourth industry, the public sector has increasingly paid more attention to search using drones and real-time monitoring, for various goals. The drones are used and researched to complete a variety of searching and monitoring missions, including search for missing persons, security, coastal patrol and monitoring, speed enforcement, highway and urban traffic monitoring, fire and wildfire monitoring, monitoring of illegal fishing in reservoirs and protest rally monitoring. Police stations, fire departments and military authorities, however, concentrate on the hardware part, so there are little research on efficient communication systems for the real-time monitoring of data collected from high-performance resolution and infrared thermal imagining cameras, and analysis programs suitable for special missions. In order to increase the efficiency of drones with the searching mission, this paper, therefore, attempts to propose an image analysis technique to increase the precision of search by producing image data suitable for searching missions, based on images obtained from drones and provide the foundation for improving relevant policies and establishing proper platforms, based on actual field cases and experiments.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.10
• /
• pp.409-418
• /
• 2022

In this paper, we propose SAAnnot-C3Pap, a semi-automatic annotation method for obtaining ground truth of a player's posture. In order to obtain ground truth about the two-dimensional joint position in the existing music domain, openpose, a two-dimensional posture estimation method, was used or manually labeled. However, automatic annotation methods such as the existing openpose have the disadvantages of showing inaccurate results even though they are fast. Therefore, this paper proposes SAAnnot-C3Pap, a semi-automated annotation method that is a compromise between the two. The proposed approach consists of three main steps: extracting postures using openpose, correcting the parts with errors among the extracted parts using supervisely, and then analyzing the results of openpose and supervisely. Perform the synchronization process. Through the proposed method, it was possible to correct the incorrect 2D joint position detection result that occurred in the openpose, solve the problem of detecting two or more people, and obtain the ground truth in the playing posture. In the experiment, we compare and analyze the results of the semi-automated annotation method openpose and the SAAnnot-C3Pap proposed in this paper. As a result of comparison, the proposed method showed improvement of posture information incorrectly collected through openpose.

• Journal of Intelligence and Information Systems
• /
• v.29 no.1
• /
• pp.143-174
• /
• 2023

With the development of deep learning technologies, Artificial Intelligence powered Optical Character Recognition (AI-OCR) has evolved to read multiple languages from various forms of images accurately. For the financial industry, where a large number of diverse documents are processed through manpower, the potential for using AI-OCR is great. In this study, we present a configuration and a design of an AI-OCR modality for use in the financial industry and discuss the platform construction with application cases. Since the use of financial domain data is prohibited under the Personal Information Protection Act, we developed a deep learning-based data generation approach and used it to train the AI-OCR models. The AI-OCR models are trained for image preprocessing, text recognition, and language processing and are configured as a microservice architected platform to process a broad variety of documents. We have demonstrated the AI-OCR platform by applying it to financial domain tasks of document sorting, document verification, and typing assistance The demonstrations confirm the increasing work efficiency and conveniences.

• Journal of Intelligence and Information Systems
• /
• v.28 no.1
• /
• pp.89-106
• /
• 2022

Recently, the amount of video data collected from smartphones, CCTVs, black boxes, and high-definition cameras has increased rapidly. According to the increasing video data, the requirements for analysis and utilization are increasing. Due to the lack of skilled manpower to analyze videos in many industries, machine learning and artificial intelligence are actively used to assist manpower. In this situation, the demand for various computer vision technologies such as object detection and tracking, action detection, emotion detection, and Re-ID also increased rapidly. However, the object detection and tracking technology has many difficulties that degrade performance, such as re-appearance after the object's departure from the video recording location, and occlusion. Accordingly, action and emotion detection models based on object detection and tracking models also have difficulties in extracting data for each object. In addition, deep learning architectures consist of various models suffer from performance degradation due to bottlenects and lack of optimization. In this study, we propose an video analysis system consists of YOLOv5 based DeepSORT object tracking model, SlowFast based action recognition model, Torchreid based Re-ID model, and AWS Rekognition which is emotion recognition service. Proposed model uses single-linkage hierarchical clustering based Re-ID and some processing method which maximize hardware throughput. It has higher accuracy than the performance of the re-identification model using simple metrics, near real-time processing performance, and prevents tracking failure due to object departure and re-emergence, occlusion, etc. By continuously linking the action and facial emotion detection results of each object to the same object, it is possible to efficiently analyze videos. The re-identification model extracts a feature vector from the bounding box of object image detected by the object tracking model for each frame, and applies the single-linkage hierarchical clustering from the past frame using the extracted feature vectors to identify the same object that failed to track. Through the above process, it is possible to re-track the same object that has failed to tracking in the case of re-appearance or occlusion after leaving the video location. As a result, action and facial emotion detection results of the newly recognized object due to the tracking fails can be linked to those of the object that appeared in the past. On the other hand, as a way to improve processing performance, we introduce Bounding Box Queue by Object and Feature Queue method that can reduce RAM memory requirements while maximizing GPU memory throughput. Also we introduce the IoF(Intersection over Face) algorithm that allows facial emotion recognized through AWS Rekognition to be linked with object tracking information. The academic significance of this study is that the two-stage re-identification model can have real-time performance even in a high-cost environment that performs action and facial emotion detection according to processing techniques without reducing the accuracy by using simple metrics to achieve real-time performance. The practical implication of this study is that in various industrial fields that require action and facial emotion detection but have many difficulties due to the fails in object tracking can analyze videos effectively through proposed model. Proposed model which has high accuracy of retrace and processing performance can be used in various fields such as intelligent monitoring, observation services and behavioral or psychological analysis services where the integration of tracking information and extracted metadata creates greate industrial and business value. In the future, in order to measure the object tracking performance more precisely, there is a need to conduct an experiment using the MOT Challenge dataset, which is data used by many international conferences. We will investigate the problem that the IoF algorithm cannot solve to develop an additional complementary algorithm. In addition, we plan to conduct additional research to apply this model to various fields' dataset related to intelligent video analysis.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.19 no.6
• /
• pp.1506-1514
• /
• 2015

In this paper, we proposed a forest fires prediction and detection system. It could provide a situation of fire prediction and detection methods using context awareness sensor. A fire occurs wide range of sensing a fire in a single camera sensor, it is difficult to detect the occurrence of a fire. In this paper, we propose an algorithm for real-time by using a temperature sensor, humidity, Co2, the flame presence information acquired and comparing the data based on multiple conditions, analyze and determine the weighting according to fire in complex situations. In addition, it is possible to differential management of intensive fire detection and prediction for required dividing the state of fire zone. Therefore we propose an algorithm to determine the prediction and detection from the fire parameters as an temperature, humidity, Co2 and the flame in real-time by using a context awareness sensor and also suggest algorithm that provide the path of fire diffusion and service the secure safety zone prediction.