• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.17 no.1
• /
• pp.31-37
• /
• 2024

In this study, we propose a machine vision-based pedestrian signal detection algorithm that operates efficiently even in computing resource-constrained environments. This algorithm demonstrates high efficiency within limited resources and is designed to minimize the impact of ambient lighting by sequentially applying HSV color space-based image processing, binarization, morphological operations, labeling, and other steps to address issues such as light glare. Particularly, this algorithm is structured in a relatively simple form to ensure smooth operation within embedded system environments, considering the limitations of computing resources. Consequently, it possesses a structure that operates reliably even in environments with low computing resources. Moreover, the proposed pedestrian signal system not only includes pedestrian signal detection capabilities but also incorporates IoT functionality, allowing wireless integration with a web server. This integration enables users to conveniently monitor and control the status of the signal system through the web server. Additionally, successful implementation has been achieved for effectively controlling 50W LED pedestrian signals. This proposed system aims to provide a rapid and efficient pedestrian signal detection and control system within resource-constrained environments, contemplating its potential applicability in real-world road scenarios. Anticipated contributions include fostering the establishment of safer and more intelligent traffic systems.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.22 no.2
• /
• pp.198-204
• /
• 2012

Tracking and recognition of robots are required for the cooperation task of robots in various environments. In the paper, a tracking control system of moving robot using stereo image processing, code-book model and fuzzy controller is proposed. First, foreground and background images are separated by using code-book model method. A candidate region is selected based on the color information in the separated foreground image and real distance of the robot is estimated from matching process of depth image that is acquired through stereo image processing. The open and close processing of image are applied and labeling according to the size of mobile robot is used to recognize the moving robot effectively. A fuzzy tracking controller using distance information and mobile information by stereo image processing is designed for effective tracking according to the movement velocity of the target robot. The proposed fuzzy tracking control method is verified through tracking experiments of mobile robots with stereo camera.

• Journal of the Korea Society of Computer and Information
• /
• v.15 no.8
• /
• pp.31-39
• /
• 2010

The paper proposes the median filter based Retinex algorithm to detect the lanes in a foggy image. Whether an input image is foggy or not is determined by analyzing the histogram in the pre-defined ROI(Region of Interest). If the image is determined as a foggy one, then it is improved by the median filter based Retinex algorithm. By replacing the Gaussian filter by the median filter in the Retinex algorithm, the processing time can be reduced and the lane features can be detected more robustly. Once the enhanced image is acquired, the binarization based on multi-threshold and the labeling operations are applied. Finally, it detects the lane information using the size and direction parameters of the detected lane features. The proposed algorithm has been evaluated by using various foggy images collected on different road conditions to prove that it detects lanes more robustly in most cases than the conventional methods.

• Resources Recycling
• /
• v.4 no.4
• /
• pp.16-21
• /
• 1995

When waste t~re/~vastmz otor oil is pyrolyzed. most of them hecome gaseous produds. and thc remaining onc, whascwelght is ahout in% oi the waste Ore, is pyrolyced residue mnstly composcd oi ca~bnn black A rescsrcll was canicrl nut loutilize lhe pyralyred residue of waste tnelwuste lnotol 011 us retnin~cing agent of asphall concrete, bescd on iolelg~r lesearchrepurl. This shows thal the properlies ol asphall concrele ~nclud~cd~ugl ah~l~tyre, sistance to Tear. temperature-v~scusilysusceptil,ilily u e g reatly improved when lhe pellellrcd hrln aI carlmn hlack usmg petroleum o ~als a hinder Iar ihe pellels isused with asphalt. The surface of the pyralyred resirh~ei s covned by ocl film and thla lnakes good comllatibllity with asphallIn order lo ulilk pyrolyzed residue as a reinforcing agenl oi lhe itsphalt concrete, various tests such as Marshnll tcsi, wheeltracking, and revelhng test has been carried out a1 KLER, Ko~ea I-lighway Coo~poration, and TCMO. Tcst lcsults satirry KSslandard, show "npmvements an the dynam~cs tab~l~lzym, d incrcase reslslance to wear at cold telnpelatule Invrsligadon wascarlied oul to sludg the possibility of soil pallul~on when pyrolyzed residue is used as a tzmioicing agenl. E~pcrimentalresulls show the rcsidue contained in thc asphall docs not cause cnv~ranma~lparlo blems.e cnv~ranma~lparlo blems.

• Convergence Security Journal
• /
• v.23 no.4
• /
• pp.61-70
• /
• 2023

Clustering is an unsupervised learning method that involves grouping data based on features such as distance metrics, using data without known labels or ground truth values. This method has the advantage of being applicable to various types of data, including images, text, and audio, without the need for labeling. Traditional clustering techniques involve applying dimensionality reduction methods or extracting specific features to perform clustering. However, with the advancement of deep learning models, research on deep clustering techniques using techniques such as autoencoders and generative adversarial networks, which represent input data as latent vectors, has emerged. In this study, we propose a deep clustering technique based on deep learning. In this approach, we use an autoencoder to transform the input data into latent vectors, and then construct a vector space according to the cluster structure and perform k-means clustering. We conducted experiments using the MNIST and Fashion-MNIST datasets in the PyTorch machine learning library as the experimental environment. The model used is a convolutional neural network-based autoencoder model. The experimental results show an accuracy of 89.42% for MNIST and 56.64% for Fashion-MNIST when k is set to 10.

• Journal of Internet Computing and Services
• /
• v.16 no.1
• /
• pp.39-46
• /
• 2015

This paper presents an automatic method to detect and count the cells from microorganism images based on mobile environments. Cell counting is an important process in the field of biological and pathological image analysis. In the past, cell counting is done manually, which is known as tedious and time consuming process. Moreover, the manual cell counting can lead inconsistent and imprecise results. Therefore, it is necessary to make an automatic method to detect and count cells from biological images to obtain accurate and consistent results. The proposed multi-step cell counting method automatically segments the cells from the image of cultivated microorganism and labels the cells by utilizing topological analysis of the segmented cells. To improve the accuracy of the cell counting, we adopt watershed algorithm in separating agglomerated cells from each other and morphological operation in enhancing the individual cell object from the image. The system is developed by considering the availability in mobile environments. Therefore, the cell images can be obtained by a mobile phone and the processed statistical data of microorganism can be delivered by mobile devices in ubiquitous smart space. From the experiments, by comparing the results between manual and the proposed automatic cell counting we can prove the efficiency of the developed system.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_1
• /
• pp.1407-1422
• /
• 2022

This study was conducted to classify tree species and assess the classification accuracy, using SE-Inception, a classification-based deep learning model. The input images of the dataset used Worldview-3 and GeoEye-1 images, and the size of the input images was divided into 10 × 10 m, 30 × 30 m, and 50 × 50 m to compare and evaluate the accuracy of classification of tree species. The label data was divided into five tree species (Pinus densiflora, Pinus koraiensis, Larix kaempferi, Abies holophylla Maxim. and Quercus) by visually interpreting the divided image, and then labeling was performed manually. The dataset constructed a total of 2,429 images, of which about 85% was used as learning data and about 15% as verification data. As a result of classification using the deep learning model, the overall accuracy of up to 78% was achieved when using the Worldview-3 image, the accuracy of up to 84% when using the GeoEye-1 image, and the classification accuracy was high performance. In particular, Quercus showed high accuracy of more than 85% in F₁ regardless of the input image size, but trees with similar spectral characteristics such as Pinus densiflora and Pinus koraiensis had many errors. Therefore, there may be limitations in extracting feature amount only with spectral information of satellite images, and classification accuracy may be improved by using images containing various pattern information such as vegetation index and Gray-Level Co-occurrence Matrix (GLCM).

• Korean Journal of Remote Sensing
• /
• v.38 no.6_2
• /
• pp.1633-1641
• /
• 2022

Urban population concentration and indiscriminate development are causing various environmental problems such as air pollution and heat island phenomena, and causing human resources to deteriorate the damage caused by natural disasters. Urban trees have been proposed as a solution to these urban problems, and actually play an important role, such as providing environmental improvement functions. Accordingly, quantitative measurement and analysis of individual trees in urban trees are required to understand the effect of trees on the urban environment. However, the complexity and diversity of urban trees have a problem of lowering the accuracy of single tree detection. Therefore, we conducted a study to effectively detect trees in Dongjak-gu using high-resolution aerial images that enable effective detection of tree objects and You Only Look Once Version 5 (YOLOv5), which showed excellent performance in object detection. Labeling guidelines for the construction of tree AI learning datasets were generated, and box annotation was performed on Dongjak-gu trees based on this. We tested various scale YOLOv5 models from the constructed dataset and adopted the optimal model to perform more efficient urban tree detection, resulting in significant results of mean Average Precision (mAP) 0.663.

• Journal of Korea Port Economic Association
• /
• v.24 no.1
• /
• pp.61-83
• /
• 2008

The meaning of "added value" refers to the contribution of the factors of production, i.e., land, labor, and capital goods, to raising the value of a product and corresponds to the incomes received by the owners of these factors. The importance of added value in service industry has been recognized as one of the critical factors to economic growth, even in logistics industry. But, it is hard to find out the previous studies providing a clear definition and framework for designing and analyzing the performance of Value Added Logistics(VAL). The purpose of this study is to define the meaning of extended VAL that extensively includes activities initiating and operating the reverse logistics under the closed-loop logistic scheme, and to suggest the framework that describes the partnerships among participants involving in operating the value added logistics. Also, in this paper, we emphasize on the need for investigation of added value logistics definition and framework based on previous academic studies, and examine various value added logistics service activities of current leading 3PL companies across the world. Finally, we suggest the analytic framework for value added logistics throughout closed-loop logistics.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2007.10a
• /
• pp.293-297
• /
• 2007

This thesis suggests a new algorithm to detects multiple moving objects using a CMODE(Correct Multiple Object DEtection) method in the color images acquired in real-time and to track the interested pedestrian using motion and hue information. The multiple objects are detected, and then shaking trees or moving cars are removed using structural characteristics and shape information of the man , the interested pedestrian can be detected, The first similarity judgment for tracking an interested pedestrian is to use the distance between the previous interested pedestrian's centroid and the present pedestrian's centroid. For the area where the first similarity is detected, three feature points are calculated using k-mean algorithm, and the second similarity is judged and tracked using the average hue value for the $3{\times}3$ area of each feature point. The zooming of camera is adjusted to track an interested pedestrian at a long distance easily and the FOV(Field of View) of camera is adjusted in case the pedestrian is not situated in the fixed range of the screen. As a experiment results, comparing the suggested CMODE method with the labeling method, an average approach rate is one fourth of labeling method, and an average detecting time is faster three times than labeling method. Even in a complex background, such as the areas where trees are shaking or cars are moving, or the area of shadows, interested pedestrian detection is showed a high detection rate of average 96.5%. The tracking of an interested pedestrian is showed high tracking rate of average 95% using the information of situation and hue, and interested pedestrian can be tracked successively through a camera FOV and zooming adjustment.