• Journal of the Semiconductor & Display Technology
• /
• v.23 no.3
• /
• pp.135-139
• /
• 2024

Public datasets, which are freely available and often labeled, play a crucial role in training object detection models in computer vision. While public datasets are effective for developing general object detection models, they may not be ideal for specialized tasks. For specific object detection needs, it is more beneficial to create and use a dataset tailored to the target object. This paper proposes a method for extracting a target-specific dataset from public datasets to develop object detection models with superior performance for the target object. This approach not only improves detection accuracy, but also reduces training data requirements and complexity. We evaluate the performance of the proposed method using the latest object detection model YOLOv10.

• Journal of information and communication convergence engineering
• /
• v.21 no.4
• /
• pp.329-336
• /
• 2023

This paper introduces an edge AI-based scene-specific object detection system for long-term traffic management, focusing on analyzing congestion and movement via cameras. It aims to balance fast processing and accuracy in traffic flow data analysis using edge computing. We adapt the YOLOv5 model, with four heads, to a scene-specific model that utilizes the fixed camera's scene-specific properties. This model selectively detects objects based on scale by blocking nodes, ensuring only objects of certain sizes are identified. A decision module then selects the most suitable object detector for each scene, enhancing inference speed without significant accuracy loss, as demonstrated in our experiments.

• The Journal of Korea Robotics Society
• /
• v.13 no.1
• /
• pp.1-7
• /
• 2018

The image-to-image translation is one of the deep learning applications using image data. In this paper, we aim at improving the performance of object transfiguration which transforms a specific object in an image into another specific object. For object transfiguration, it is required to transform only the target object and maintain background images. In the existing results, however, it is observed that other parts in the image are also transformed. In this paper, we have focused on the structure of artificial neural networks that are frequently used in the existing methods and have improved the performance by adding constraints to the exiting structure. We also propose the advanced structure that combines the existing structures to maintain their advantages and complement their drawbacks. The effectiveness of the proposed methods are shown in experimental results.

• Proceedings of the ESK Conference
• /
• 1996.10a
• /
• pp.197-205
• /
• 1996

Unification of neural network with a hierarchical pattern recognition is presented for recognizing large set of objects. A two-step identification procedure is developed for pattern recognition: coarse and fine identification. The coarse identification is designed for finding a class of object while the fine identification procedure is to identify a specific object. During the training phase a course neural network is trained for clustering larger set of reference objects into a number of groups. For training a fine neural network, expert neural network is also trained to identify a specific object within a group. The presented idea can be interpreted as two step identification. Experimental results are given to verify the proposed methodology.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.31 no.3
• /
• pp.1-7
• /
• 2008

This paper proposes an object-oriented software development guidance and an evaluation index for the productivity related to EJB(Enterprise JavaBeans). EJB is a known successful standard model for LSDO(Large Size Distributed Object). However, there is no comparison research about the performance of EJB 2.0 and 3.0 with same identical platform. Quantitative analysis is supported as a part of LOC(Line Of Code) analysis. There is a limit to develop the updated software with no the specific evaluating index for the productivity of software. This work proposes an specific index for evaluating the productivity of new version EJB on a platform. Base on the result, the specific guidance of the developing software is obtained.

• Korean Journal of Computational Design and Engineering
• /
• v.21 no.3
• /
• pp.325-333
• /
• 2016

This paper aims to classify objects and their properties represented in Korea Building Act sentences for applying to BIM-enabled automated code compliance checking task. In order to conduct automated code compliance checking, it is necessary to develop translation process of converting the building act sentences into computer-executable forms. However, since Korea building act sentences are written in natural language, some of requirements are ambiguous to translate explicitly. In this regard, the building act sentences regarding building permit requirements are analyzed focusing on the regulation-specific objects and related properties representation from noun phrases within the scope of this paper. From 1977 building act sentences and attached reference regulations, 1200 regulation-specific objects and about 220 related properties are extracted and classified. In the application for the classification, object-property database is implemented and some of application using the database and the regulation-specific classification is suggested to support to generate rule set written in computable codes.

• Journal of the Korea Society of Computer and Information
• /
• v.13 no.2
• /
• pp.135-141
• /
• 2008

In this paper, we propose colored-hear recognition that distinguishing feature of synesthesia for human sensing by shared vision and specific sense of hearing. We perceived what potential influence of human's structured object recognition by visual analysis through the camera, So we've studied how to make blind persons can feel similar vision of real object. First of all, object boundaries are detected in the image data representing a specific scene. Then, four specific features such as object location in the image focus, feeling of average color, distance information of each object, and object area are extracted from picture. Finally, mapping these features to the audition factors. The audition factors are used to recognize vision for blind persons. Proposed colored-hear transformation for recognition can get fast and detail perception, and can be transmit information for sense at the same time. Thus, we were get a food result when applied this concepts to blind person's case of image recognition.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.4
• /
• pp.1050-1061
• /
• 2000

This paper presents a new object model, LOM(Lightweight Object Model) and an implementation method for the distributed trader in heterogeneous distributed computing environment including mobile network. Trader is third party object that enables clients to find suitable servers, which provide the most appropriate services to client in distributed environment including dynamic reconfiguration of services and servers. Trading service requires simpler and more specific object model than genetic object models which provide richer multimedia data types and semantic characteristics with complex data structures. LOM supports a new reference attribute type instead of the relationship, inheritance and composite attribute types of the general object oriented models and so LOM has simple data structures. Also in LOM, the modelling step includes specifying of the information about users and the access right to objects for security in the mobile environment and development of the distributed storage for trading service. Also, we propose and implementation method of the distributed trader, which integrates the LOM-information object model and the OMG (object Management Group) computational object model.

• Journal of the Korea Society of Computer and Information
• /
• v.24 no.8
• /
• pp.19-28
• /
• 2019

In this paper, an online video summarization algorithm based on multiple object detection is proposed. As crime has been on the rise due to the recent rapid urbanization, the people's appetite for safety has been growing and the installation of surveillance cameras such as a closed-circuit television(CCTV) has been increasing in many cities. However, it takes a lot of time and labor to retrieve and analyze a huge amount of video data from numerous CCTVs. As a result, there is an increasing demand for intelligent video recognition systems that can automatically detect and summarize various events occurring on CCTVs. Video summarization is a method of generating synopsis video of a long time original video so that users can watch it in a short time. The proposed video summarization method can be divided into two stages. The object extraction step detects a specific object in the video and extracts a specific object desired by the user. The video summary step creates a final synopsis video based on the objects extracted in the previous object extraction step. While the existed methods do not consider the interaction between objects from the original video when generating the synopsis video, in the proposed method, new object clustering algorithm can effectively maintain interaction between objects in original video in synopsis video. This paper also proposed an online optimization method that can efficiently summarize the large number of objects appearing in long-time videos. Finally, Experimental results show that the performance of the proposed method is superior to that of the existing video synopsis algorithm.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.8
• /
• pp.1117-1123
• /
• 2021

Object recognition is to identify the location and class of a specific object by analyzing the given image when a specific image is input. One of the fields in which object recognition technology is actively applied in recent years is autonomous vehicles, and this paper describes the trend of image-based object recognition artificial intelligence technology in autonomous vehicles. The image-based object detection algorithm has recently been narrowed down to two methods (a single-step detection method and a two-step detection method), and we will analyze and organize them around this. The advantages and disadvantages of the two detection methods are analyzed and presented, and the YOLO/SSD algorithm belonging to the single-step detection method and the R-CNN/Faster R-CNN algorithm belonging to the two-step detection method are analyzed and described. This will allow the algorithms suitable for each object recognition application required for autonomous driving to be selectively selected and R&D.