• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.4
• /
• pp.473-478
• /
• 2005

The gastric cancer takes the great part of the cancer occurrence and the mortality from cancer in Korea, and the early detection of gastric cancer is very important in the treatment and convalescence. This paper. for the early detection of gastric cancer, Proposes the analysis system of endoscopic image of the stomach that detects the abnormal region by using the change of color in the image and provides the surface tissue information to the detector. While the advanced inflammation and the cancer may be easily detected, the early inflammation and the cancer have a difficulty in detection and require the more attention lot detection. This paper, at first, converts the endoscopic image to the Image of IHb(Index of Hemoglobin) model and removes noises incurred by illumination, and next, automatically detects the regions suspected as cancer and provides the related information to the detector, or provides the surface tissue information for the regions appointed by the detector. This paper does not intend to provide the final diagnosis of the detected abnormal regions as gastric cancer, but provides the supplementary mean that reduces the load and mistaken diagnosis of the detector by automatically detecting the abnormal regions being not easily detected by human eyes and providing the additional information for the diagnosis. The experiments using practical endoscopic images for performance evaluation showed that the proposed system is effective in the analysis of endoscopic image of the stomach.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.21 no.6
• /
• pp.749-754
• /
• 2011

In this study, we propose the design of optimized pRBFNNs-based face recognition system using two-dimensional Image and ASM algorithm. usually the existing 2 dimensional face recognition methods have the effects of the scale change of the image, position variation or the backgrounds of an image. In this paper, the face region information obtained from the detected face region is used for the compensation of these defects. In this paper, we use a CCD camera to obtain a picture frame directly. By using histogram equalization method, we can partially enhance the distorted image influenced by natural as well as artificial illumination. AdaBoost algorithm is used for the detection of face image between face and non-face image area. We can butt up personal profile by extracting the both face contour and shape using ASM(Active Shape Model) and then reduce dimension of image data using PCA. The proposed pRBFNNs consists of three functional modules such as the condition part, the conclusion part, and the inference part. In the condition part of fuzzy rules, input space is partitioned with Fuzzy C-Means clustering. In the conclusion part of rules, the connection weight of RBFNNs is represented as three kinds of polynomials such as constant, linear, and quadratic. The essential design parameters (including learning rate, momentum coefficient and fuzzification coefficient) of the networks are optimized by means of Differential Evolution. The proposed pRBFNNs are applied to real-time face image database and then demonstrated from viewpoint of the output performance and recognition rate.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.22 no.3
• /
• pp.451-457
• /
• 2018

We propose an automatic detection and classification system of sewer damage database based on artificial intelligence and deep learning. In order to optimize the performance, we implemented a robust system against various environmental variations such as illumination and shadow changes. In our proposed system, a crack detection and damage classification method using a deep learning based Convolutional Neural Network (CNN) is implemented. For optimal results, 9,941 CCTV images with $256{\times}256$ pixel resolution were used for machine learning on the damaged area based on the CNN model. As a result, the recognition rate of 98.76% was obtained. Total of 646 images of $720{\times}480$ pixel resolution were extracted from various sewage DB for performance evaluation. Proposed system presents the optimal recognition rate for the automatic detection and classification of damage in the sewer DB constructed in various environments.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.26 no.2
• /
• pp.129-152
• /
• 2024

In the application of deep learning object detection via CCTV in tunnels, a large number of false positive detections occur due to the poor environmental conditions of tunnels, such as low illumination and severe perspective effect. This problem directly impacts the reliability of the tunnel CCTV-based accident detection system reliant on object detection performance. Hence, it is necessary to reduce the number of false positive detections while also enhancing the number of true positive detections. Based on a deep learning object detection model, this paper proposes a false positive data training method that not only reduces false positives but also improves true positive detection performance through retraining of false positive data. This paper's false positive data training method is based on the following steps: initial training of a training dataset - inference of a validation dataset - correction of false positive data and dataset composition - addition to the training dataset and retraining. In this paper, experiments were conducted to verify the performance of this method. First, the optimal hyperparameters of the deep learning object detection model to be applied in this experiment were determined through previous experiments. Then, in this experiment, training image format was determined, and experiments were conducted sequentially to check the long-term performance improvement through retraining of repeated false detection datasets. As a result, in the first experiment, it was found that the inclusion of the background in the inferred image was more advantageous for object detection performance than the removal of the background excluding the object. In the second experiment, it was found that retraining by accumulating false positives from each level of retraining was more advantageous than retraining independently for each level of retraining in terms of continuous improvement of object detection performance. After retraining the false positive data with the method determined in the two experiments, the car object class showed excellent inference performance with an AP value of 0.95 or higher after the first retraining, and by the fifth retraining, the inference performance was improved by about 1.06 times compared to the initial inference. And the person object class continued to improve its inference performance as retraining progressed, and by the 18th retraining, it showed that it could self-improve its inference performance by more than 2.3 times compared to the initial inference.

• 대한공업교육학회지
• /
• v.34 no.1
• /
• pp.87-112
• /
• 2009

The aim of this research is to suggest the model of reconstruction and remodelling of S high school of broadcast for converting into a specialized one in the section of broadcast industry. The method of reconstruction of the laboratories by each department is as follows. First, the names of laboratories, ground plans, and perspective drawings are designed according to the principle of constructionism. Second, each department is recommended to install 7 laboratories. For the department of broadcasting facilities, a laboratory of processing broadcasting machines, a laboratory of broadcasting machines, and a laboratory of screen picture machines are needed. For the department of broadcasting, electricity, and communication, a laboratory of image and sound edition, a laboratory of broadcasting and communication, a laboratory of broadcasting system, and a laboratory of stage illumination. For the department of broadcasting stage construction, a laboratory of broadcasting stage building, a laboratory of broadcasting stage design, a laboratory of broadcasting stage decoration and etc. Third, an office room is designed by department in the form of seminar room and places for practice is formed by department to be a laboratory. Fourth, a room for industrial-educational cooperation is installed by department, by which education of industrial-educational cooperation is carried out harmoniously. And one example of a remodeled laboratory by one department is shown with laboratory name, ground plan, perspective drawing, and running method. Especially employment of teachers working both school and business, professional club which is concerned with broadcasting technique, and attraction of enterprise in school is suggested as important elements to manage broadcasting laboratories efficiently. Finally, it is desirable for us to attract private capital, that is, Build-Transfer-Lease(BTL) because much budget is required for purchasing the broadcasting equipments and remodelling the laboratories to succeed as specialized high school on broadcasting section.

• Journal of Mushroom
• /
• v.16 no.3
• /
• pp.155-161
• /
• 2018

In this study, we analyze the growth environment using smart farm technology in order to develop the optimal growth model for the precision cultivation of the bottle-grown oyster mushroom 'Suhan'. Experimental farmers used $88m^2$ of bed area, 2 rows and 5 columns of shelf shape, 5 hp refrigerator, 100T of sandwich panel for insulation, 2 ultrasonic humidifiers, 12 kW of heating, and 5,000 bottles for cultivation. Data on parameters such as temperature, humidity, carbon dioxide concentration, and illumination, which directly affect mushroom growth, were collected from the environmental sensor part installed at the oyster mushroom cultivator and analyzed. It was found that the initial temperature at the time of granulation was $22^{\circ}C$ after the scraping, and the mushroom was produced and maintained at about $25^{\circ}C$ until the bottle was flipped. On fruiting body formation, mushrooms were harvested while maintaining the temperature between $13^{\circ}C$ and $15^{\circ}C$. Humidity was approximately 100% throughout the growth stage. Carbon dioxide concentration gradually increased until 3 days after the beginning of cultivation, and then increased rapidly to approximately 2,600 ppm. From the 6th day, $CO_2$ concentration was gradually decreased through ventilation and maintained at 1,000 ppm during the harvest. Light was not provided at the initial stage of oyster mushroom cultivation. On the $3^{rd}$ and $4^{th}$ day, mushrooms were irradiated by 17 lux light. Subsequently, the light intensity was increased to 115-120 lux as the growth progressed. Fruiting body characteristics of 'Suhan' cultivated in a farmhouse were as follows: Pileus diameter was 30.9 mm and thickness was 4.5 mm; stipe thickness was 11.0 mm and length was 76.0 mm; stipe and pileus hardness was 0.8 g/mm and 2.8 g/mm, respectively; L values of the stipe and pileus were 79.9 and 52.3, respectively. The fruiting body yield was 160.2 g/850 ml, and the individual weight was 12.8 g/10 unit.

• Journal of Mushroom
• /
• v.17 no.3
• /
• pp.119-125
• /
• 2019

This study aims to report the results for the analysis of the growth environment by applying smart farm technology to "Chunchu No 2" farmers in order to develop an optimal growth model for precision cultivation of bottle-grown oyster mushrooms. The temperature, humidity, carbon dioxide concentration, and illumination data were collected and analyzed using an environmental sensor installed to obtain growth environment data from the oyster mushroom cultivator. Analysis of the collected temperature data revealed that the temperature at the time of granulation was $19.5^{\circ}C$ after scraping, and the mushroom was generated and maintained at about $21^{\circ}C$ until the bottle was flipped. When the fruiting body grew and approached harvest time, mushrooms were harvested while maintaining the temperature between $14^{\circ}C$ and $18^{\circ}C$. The humidity was maintained at almost 100% during the complete growth stage. Carbon dioxide concentration gradually increased until 3 days after the beginning of cultivation, and then increased rapidly to almost 5,500 ppm. From the 6th day, carbon dioxide concentration was gradually decreased through ventilation and was maintained at 1,600 ppm during harvest. Light intensity of 8 lux was irradiated up to day 6 after seeding, and growth was then continued while periodically irradiating 4 lux light. The fruiting body characteristics of "Chunchu No 2" cultivated in the farmhouse were as follows: pileus diameter of 26.5 mm and thickness of 4.9 mm, stipe thickness of 8.9 mm, and length of 68.7 mm. The fruiting body yield was 166.8 g/850 ml, and the individual weight was 12.8 g/10 units.

• Journal of Korean Society of Environmental Engineers
• /
• v.30 no.9
• /
• pp.955-960
• /
• 2008

We investigated and estimated at the characteristics of decomposition and mineralization of benomyl using a design of experiment(DOE) based on the general factorial design in an E-beam process, and also the main factors(variables) with benomyl concentration(X$_1$) and E-beam irradiation(X$_2$) which consisted of 5 levels in each factor was set up to estimate the prediction model and the optimization conditions. At frist, the benomyl in all treatment combinations except 17 and 18 trials was almost degraded and the difference in the decomposition of benomyl in the 3 blocks was not significant(p > 0.05, one-way ANOVA). However, the % of benomyl mineralization was 46%(block 1), 36.7%(block 2) and 22%(block 3) and showed the significant difference of the % that between each block(p < 0.05). The linear regression equations of benomyl mineralization in each block were also estimated as followed; block 1(Y$_1$ = 0.024X$_1$ + 34.1(R$^2$ = 0.929)), block 2(Y$_2$ = 0.026X$_2$ + 23.1(R$^2$ = 0.976)) and block 3(Y$_3$ = 0.034X$_3$ + 6.2(R$^2$ = 0.98)). The normality of benomyl mineralization obtained from Anderson-Darling test in all treatment conditions was satisfied(p > 0.05). The results of prediction model and optimization point using the canonical analysis in order to obtain the optimal operation conditions were Y = 39.96 - 9.36X$_1$ + 0.03X$_2$ - 10.67X$_1{^2}$ - 0.001X$_2{^2}$ + 0.011X$_1$X$_2$(R$^2$ = 96.3%, Adjusted R$^2$ = 94.8%) and 57.3% at 0.55 mg/L and 950 Gy, respectively. A Microtox test using V. fischeri showed that the toxicity, expressed as the inhibition(%), was reduced almost completely after an E-beam irradiation, whereas the inhibition(%) for 0.5 mg/L, 1 mg/L and 1.5 mg/L was 10.25%, 20.14% and 26.2% in the initial reactions in the absence of an E-beam illumination.