• Journal of the Korea Institute of Building Construction
• /
• v.23 no.2
• /
• pp.165-173
• /
• 2023

This research proposed a model for automatically monitoring the quality of insulation adhesive application in external insulation construction. Upon case implementation, the area segmentation model demonstrated a 92.3% accuracy, while the area and distance calculation accuracies of the proposed model were 98.8% and 96.7%, respectively. These findings suggest that the model can effectively prevent the most common insulation defect, insulation failure, while simultaneously minimizing the need for on-site supervisory personnel during external insulation construction. This, in turn, contributes to the enhancement of the external insulation system. Moving forward, we plan to gather construction images of various external insulation methods to refine the image segmentation model's performance and develop a model capable of automatically monitoring scenarios with a considerable number of insulation materials in the image.

• Korean Journal of Community Nutrition
• /
• v.10 no.6
• /
• pp.905-919
• /
• 2005

This study raised the necessity of developing performance indicators for measuring the management efficiency and effectiveness of school food service, and as a means of helping its implementation, a balanced score card (BSC) approach developed by Norton and Kaplan was adopted. This study established BSC in seven phases through literature: Phase 1 Defining a school food service and the scope of working activities, Phase 2 Establishing the vision of a school food service, Phase 3 Setting strategic goals, Phase 4 Identifying critical success factors (CSFs), Phase 5 Developing Key Performance Indicators (KPIs), Phase 6 Extracting cause and effect relationship, and Phase 7 Completing a preliminary BSC. The preliminary BSC was fumed into a survey, which was administered to food service related people working at the Office of Education and School Food Service including 16 offices,209 dietitians, 48 school administrators both from self-operated and contract-managed, and 9 experts in areas related to school food service. They were asked questions about strategies from 4 different perspectives,12 CSFs, 39 KPls, and the cause and effect relationships among them. As a result, among the CSFs based on 4 different perspectives, all factors other than ' zero sum on profit/loss ' from the financial perspective turned out to be valid. In terms of KPIs, manufacturing cost percentages, casualty loss count/reduction rates, school foodervice participation rates, and sales goal achievement rates were found to be valid from the financial perspective, while student satisfaction index, faculty satisfaction index, leftover ratio, nutrition educational performance count, index of evaluating nutrition education, customer claim count/reduction rate, handling customer claim count/reduction rate, and parent satisfaction index were found to be valid from the customers' perspective. Besides, nutritional requirement sufficient ratio, nutritional management score, food poisoning outbreak count, employee safety accident count, sanitary inspection assessment index, meals per labor hour (productivity index), computerization ratio, operational management index, and purchase management assessment index were also found to be valid from the perspective of internal business processes. From the perspective of innovation and learning, employee turnover ratio/rate of absenteeism, annual education and training count, employee satisfaction index, human resource management assessment index, annual menu-related customer feedback, food service information index for employees and parents/schools were also found to be valid. The significance of this study is to present indices for measuring overall performance of school lunch food service operations without putting any limitation on types of school food service management, and to help correctly assess the contribution of the current types of school food service management to schools and students. (Korean J Community Nutrition 10(6) : $905\∼919$, 2005)

• The Journal of the Korea Contents Association
• /
• v.17 no.11
• /
• pp.303-311
• /
• 2017

Safety accidents in railway work continue to increase every year; Engineer's negligence, trackside worker's sensory deprivation and signalman's mistake are the main reasons of such incidents. We consider this problem by far the most urgent matter in railway work because of its steady increase and risk of taking a person's life. Based on that account, a new alarm system has developed, that is called Railway Access Alarm System, to allow railway workers to sense the access of trains with not only vision, but also hearing. The detector device of this system is installed on both sides of the track locating 1.5km from the workplace. When the train enters the place, the detector device can sense the entering, sending the detect sign of train to the alarm unit, then the alarm unit warns the workers by the LED lighting and sirens. This system has several advantages compared to previous systems. First, it recognizes the train at a long distance. Secondly, there is no need for wiring work since it is a wireless system. At last, the system works by rechargeable batteries and solar charger so that it is installed in the work places where there is no external power supply. Moreover, it is proven that the system is 100% reliable by the successful on-the spot inspection evaluating the capability.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.24 no.3
• /
• pp.271-278
• /
• 2014

In spite of that stop line and crosswalk should be aware of the most basic objects in transportation system, its features extracted are very limited. In addition to image-based recognition technology, laser and RF, GPS/INS recognition technology, it is difficult to recognize. For this reason, the limited research in this area has been done. In this paper, the algorithm to recognize the stop line and crosswalk is designed and implemented using image-based recognition technology with the images input through a vision sensor. This algorithm consists of three functions.; One is to select the area, in advance, needed for feature extraction in order to speed up the data processing, 'Region of Interest', another is to process the images only that white color is detected more than a certain proportion in order to remove the unnecessary operation, 'Color Pattern Inspection', the other is 'Feature Extraction and Recognition', which is to extract the edge features and compare this to the previously-modeled one to identify the stop line and crosswalk. For this, especially by using case based feature comparison algorithm, it can identify either both stop line and crosswalk exist or just one exists. Also the proposed algorithm is to develop existing researches by comparing and analysing effect of in-vehicle camera installation and changes in recognition rate of distance estimation and various constraints such as backlight and shadow.

• Explosives and Blasting
• /
• v.22 no.3
• /
• pp.57-64
• /
• 2004

There are two major types of underwater blasting at Korea, bridges and harbor construction work. Pier blasting for lay the foundation bridges construction is used dry excavation working (drilling and charging) after pump out water and then fire pump in water that is same as bench blasting. In contrast, underwater blasting for harbor construction and increase of harbor load depth is used to barge with digging equipment that is in oder to drilling on the surface and blasting work(charge, hook-up) under water. Thus, there are need to special concern such as charge method and hook-up method different from tunnel blasting work and bench blasting work. If do not use special concern breaks out dead pressure and mis fire because of there are so many difficult condition such as water pressure, obstruct field of vision. In this study underwater blasting at Busan Harbor Construction have consider with special concern that is plastic pipe charge method used to MegaMITE I and specialized buoy hook- up method make far initial system detonate on the surface used to TLD. The results is designed blast pattern charge per delay effect an inspection of verify between predict velocity and measure velocity. minimized break out mis fire consideration charge method, hook up method. According to result best underwater blasting design is 105mm drilling dia, MeGAMITE II, HiNLL Plus(non electric detonator).

• Journal of Internet Computing and Services
• /
• v.23 no.3
• /
• pp.35-45
• /
• 2022

Even in recent years, treatment of first-aid patients is still often delayed due to a shortage of medical resources in marginalized areas. Research on automating the analysis of medical data to solve the problems of inaccessibility for medical services and shortage of medical personnel is ongoing. Computer vision-based medical inspection automation requires a lot of cost in data collection and labeling for training purposes. These problems stand out in the works of classifying lesion that are rare, or pathological features and pathogenesis that are difficult to clearly define visually. Anomaly detection is attracting as a method that can significantly reduce the cost of data collection by adopting an unsupervised learning strategy. In this paper, we propose methods for detecting abnormal images on chest X-RAY images as follows based on existing anomaly detection techniques. (1) Normalize the brightness range of medical images resampled as optimal resolution. (2) Some feature vectors with high representative power are selected in set of patch features extracted as intermediate-level from lesion-free images. (3) Measure the difference from the feature vectors of lesion-free data selected based on the nearest neighbor search algorithm. The proposed system can simultaneously perform anomaly classification and localization for each image. In this paper, the anomaly detection performance of the proposed system for chest X-RAY images of PA projection is measured and presented by detailed conditions. We demonstrate effect of anomaly detection for medical images by showing 0.705 classification AUROC for random subset extracted from the PadChest dataset. The proposed system can be usefully used to improve the clinical diagnosis workflow of medical institutions, and can effectively support early diagnosis in medically poor area.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.5
• /
• pp.211-220
• /
• 2022

Recently, research on smart factories triggered by the 4th industrial revolution is being actively conducted. Accordingly, the manufacturing industry is conducting various studies to improve productivity and quality based on deep learning technology with robust performance. This paper is a study on the method of detecting tire surface defects in the visual inspection stage of the tire manufacturing process, and introduces a tire surface defect detection method using a depth image acquired through a 3D camera. The tire surface depth image dealt with in this study has the problem of low contrast caused by the shallow depth of the tire surface and the difference in the reference depth value due to the data acquisition environment. And due to the nature of the manufacturing industry, algorithms with performance that can be processed in real time along with detection performance is required. Therefore, in this paper, we studied a method to normalize the depth image through relatively simple methods so that the tire surface defect detection algorithm does not consist of a complex algorithm pipeline. and conducted a comparative experiment between the general normalization method and the normalization method suggested in this paper using YOLO V3, which could satisfy both detection performance and speed. As a result of the experiment, it is confirmed that the normalization method proposed in this paper improved performance by about 7% based on mAP 0.5, and the method proposed in this paper is effective.

• Journal of IKEEE
• /
• v.25 no.3
• /
• pp.493-500
• /
• 2021

In this paper, we propose the development of deep learning structure to improve quality of polygonal containers. The deep learning structure consists of a convolution layer, a bottleneck layer, a fully connect layer, and a softmax layer. The convolution layer is a layer that obtains a feature image by performing a convolution 3x3 operation on the input image or the feature image of the previous layer with several feature filters. The bottleneck layer selects only the optimal features among the features on the feature image extracted through the convolution layer, reduces the channel to a convolution 1x1 ReLU, and performs a convolution 3x3 ReLU. The global average pooling operation performed after going through the bottleneck layer reduces the size of the feature image by selecting only the optimal features among the features of the feature image extracted through the convolution layer. The fully connect layer outputs the output data through 6 fully connect layers. The softmax layer multiplies and multiplies the value between the value of the input layer node and the target node to be calculated, and converts it into a value between 0 and 1 through an activation function. After the learning is completed, the recognition process classifies non-circular glass bottles by performing image acquisition using a camera, measuring position detection, and non-circular glass bottle classification using deep learning as in the learning process. In order to evaluate the performance of the deep learning structure to improve quality of polygonal containers, as a result of an experiment at an authorized testing institute, it was calculated to be at the same level as the world's highest level with 99% good/defective discrimination accuracy. Inspection time averaged 1.7 seconds, which was calculated within the operating time standards of production processes using non-circular machine vision systems. Therefore, the effectiveness of the performance of the deep learning structure to improve quality of polygonal containers proposed in this paper was proven.

• Journal of Food Hygiene and Safety
• /
• v.34 no.1
• /
• pp.1-12
• /
• 2019

The health benefits associated with consumption of fresh produce have been clearly demonstrated and encouraged by international nutrition and health authorities. However, since fresh produce is usually minimally processed, increased consumption of fresh fruits and vegetables has also led to a simultaneous escalation of foodborne illness cases. According to the report by the World Health Organization (WHO), 1 in 10 people suffer from foodborne diseases and 420,000 die every year globally. In comparison to other processed foods, fresh produce can be easily contaminated by various routes at different points in the supply chain from farm to fork. This review is focused on the identification and characterization of possible sources of foodborne illnesses from chemical, biological, and physical hazards and the applicable methodologies to detect potential contaminants. Agro-chemicals (pesticides, fungicides and herbicides), natural toxins (mycotoxins and plant toxins), and heavy metals (mercury and cadmium) are the main sources of chemical hazards, which can be detected by several methods including chromatography and nano-techniques based on nanostructured materials such as noble metal nanoparticles (NMPs), quantum dots (QDs) and magnetic nanoparticles or nanotube. However, the diversity of chemical structures complicates the establishment of one standard method to differentiate the variety of chemical compounds. In addition, fresh fruits and vegetables contain high nutrient contents and moisture, which promote the growth of unwanted microorganisms including bacterial pathogens (Salmonella, E. coli O157: H7, Shigella, Listeria monocytogenes, and Bacillus cereus) and non-bacterial pathogens (norovirus and parasites). In order to detect specific pathogens in fresh produce, methods based on molecular biology such as PCR and immunology are commonly used. Finally, physical hazards including contamination by glass, metal, and gravel in food can cause serious injuries to customers. In order to decrease physical hazards, vision systems such as X-ray inspection have been adopted to detect physical contaminants in food, while exceptional handling skills by food production employees are required to prevent additional contamination.