• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.55 no.4
• /
• pp.285-293
• /
• 2019

The mesh selectivity of hairtail (Trichiurus lepturus) caught by coastal drift gill net was examined in field experiments with three different mesh sizes (45, 50 and 55 mm) from October to November, 2013 in the coastal areas of south-west of Jeju province. The mesh selectivity tests were conducted with the experimental net to be set middle part of conventional driftnets. The mesh selectivity tests were carried out the total of four times. The selectivity curve was estimated by the Kitahara's and Fujimori's method. In the results, the catch number of hairtail was 653 (125.8 kg) and occupied 34.8% in total catches weight. The optimal mesh size for 50% selection on the minimum landing size (180 mm, AL) and the first maturity size (260 mm, AL) of hairtail were estimated as 47.2 mm and 64.5 mm by master selectivity curves, respectively.

• Journal of Korea Multimedia Society
• /
• v.24 no.7
• /
• pp.849-859
• /
• 2021

Prostate cancer (PCa) is a fatal disease that occurs in men. In general, PCa cells are found in the prostate gland. Early diagnosis is the key to prevent the spreading of cancers to other parts of the body. In this case, deep learning-based systems can detect and distinguish histological patterns in microscopy images. The histological grades used for the analysis were benign, grade 3, grade 4, and grade 5. In this study, we attempt to use transfer learning and fine-tuning methods as well as different model architectures to develop and compare the models. We implemented MobileNet, ResNet50, and DenseNet121 models and used three different strategies of freezing layers techniques of fine-tuning, to get various pre-trained weights to improve accuracy. Finally, transfer learning using MobileNet with the half-layer frozen showed the best results among the nine models, and 90% accuracy was obtained on the test data set.

• Transactions of Materials Processing
• /
• v.31 no.4
• /
• pp.229-239
• /
• 2022

Convolution Neural Network(CNN) is a class of deep learning algorithms and can be used for image analysis. In particular, it has excellent performance in finding the pattern of images. Therefore, CNN is commonly applied for recognizing, learning and classifying images. In this study, the surface defect classification performance of Al 6061 extruded material using CNN-based algorithms were compared and evaluated. First, the data collection criteria were suggested and a total of 2,024 datasets were prepared. And they were randomly classified into 1,417 learning data and 607 evaluation data. After that, the size and quality of the training data set were improved using data augmentation techniques to increase the performance of deep learning. The CNN-based algorithms used in this study were VGGNet-16, VGGNet-19, ResNet-50 and DenseNet-121. The evaluation of the defect classification performance was made by comparing the accuracy, loss, and learning speed using verification data. The DenseNet-121 algorithm showed better performance than other algorithms with an accuracy of 99.13% and a loss value of 0.037. This was due to the structural characteristics of the DenseNet model, and the information loss was reduced by acquiring information from all previous layers for image identification in this algorithm. Based on the above results, the possibility of machine vision application of CNN-based model for the surface defect classification of Al extruded materials was also discussed.

• Journal of information and communication convergence engineering
• /
• v.20 no.3
• /
• pp.219-225
• /
• 2022

In this study, we propose training a support vector machine (SVM) model on top of deep networks for detecting Covid-19 from chest X-ray images. We started by gathering a real chest X-ray image dataset, including positive Covid-19, normal cases, and other lung diseases not caused by Covid-19. Instead of training deep networks from scratch, we fine-tuned recent pre-trained deep network models, such as DenseNet121, MobileNet v2, Inception v3, Xception, ResNet50, VGG16, and VGG19, to classify chest X-ray images into one of three classes (Covid-19, normal, and other lung). We propose training an SVM model on top of deep networks to perform a nonlinear combination of deep network outputs, improving classification over any single deep network. The empirical test results on the real chest X-ray image dataset show that deep network models, with an exception of ResNet50 with 82.44%, provide an accuracy of at least 92% on the test set. The proposed SVM on top of the deep network achieved the highest accuracy of 96.16%.

• Journal of Internet Computing and Services
• /
• v.20 no.4
• /
• pp.91-102
• /
• 2019

In autonomous driving systems, the ability to classify pedestrians in images captured by cameras is very important for pedestrian safety. In the past, after extracting features of pedestrians with HOG(Histogram of Oriented Gradients) or SIFT(Scale-Invariant Feature Transform), people classified them using SVM(Support Vector Machine). However, extracting pedestrian characteristics in such a handcrafted manner has many limitations. Therefore, this paper proposes a method to classify pedestrians reliably and effectively using CNN's(Convolutional Neural Network) deep features and transfer learning. We have experimented with both the fixed feature extractor and the fine-tuning methods, which are two representative transfer learning techniques. Particularly, in the fine-tuning method, we have added a new scheme, called M-Fine(Modified Fine-tuning), which divideslayers into transferred parts and non-transferred parts in three different sizes, and adjusts weights only for layers belonging to non-transferred parts. Experiments on INRIA Person data set with five CNN models(VGGNet, DenseNet, Inception V3, Xception, and MobileNet) showed that CNN's deep features perform better than handcrafted features such as HOG and SIFT, and that the accuracy of Xception (threshold = 0.5) isthe highest at 99.61%. MobileNet, which achieved similar performance to Xception and learned 80% fewer parameters, was the best in terms of efficiency. Among the three transfer learning schemes tested above, the performance of the fine-tuning method was the best. The performance of the M-Fine method was comparable to or slightly lower than that of the fine-tuningmethod, but higher than that of the fixed feature extractor method.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.34 no.4
• /
• pp.386-392
• /
• 1998

• Korean Journal of Ichthyology
• /
• v.29 no.1
• /
• pp.41-51
• /
• 2017

To understand the application in farm for the fish aquaculture, we investigated biological and pathological traits on red sea bream Pagrus major which were reared in each copper-alloy net cage and the synthetic fiber net cage for 9 months. Two groups of cage were made and set in Yokji-eup, Tongyoung, Gyeongsangnam-do in size of 25 m in diameter and 10 m of depth. Survival rate of the red sea bream in the rearing copper-alloy net cage and synthetic fiber cage showed 99.75% and 99.70% respectively, there was no significant difference. Daily weight growth rate in each net was shown to 2.13 g/day and 1.65 g/day. Health analysis by blood composition analysis showed a favorable result in the copper-alloy net cage rather than in the synthetic fiber net. Bioaccumulation of heavy metal such as Cu and Zn especially in gonad was higher than other organ. Bioaccumulation of Cu and Zn in the muscle was lower compared to the permitted standard for food safety. Pathogenic infection test discovered Microcotyle tai for parasite, V. alginolyticus and other five species for bacteria. But there was a little bit difference of bacteria infection in copper-alloy net cage and copper-alloy net cage is expected to be has antibacterial effect. Thus, copper-alloy net cage can be applied to farm considering its system stability, recycling, antibiosis and food safety.

• Korean Journal of Radiology
• /
• v.20 no.8
• /
• pp.1275-1284
• /
• 2019

Objective: To develop algorithms using convolutional neural networks (CNNs) for automatic segmentation of acute ischemic lesions on diffusion-weighted imaging (DWI) and compare them with conventional algorithms, including a thresholding-based segmentation. Materials and Methods: Between September 2005 and August 2015, 429 patients presenting with acute cerebral ischemia (training:validation:test set = 246:89:94) were retrospectively enrolled in this study, which was performed under Institutional Review Board approval. Ground truth segmentations for acute ischemic lesions on DWI were manually drawn under the consensus of two expert radiologists. CNN algorithms were developed using two-dimensional U-Net with squeeze-and-excitation blocks (U-Net) and a DenseNet with squeeze-and-excitation blocks (DenseNet) with squeeze-and-excitation operations for automatic segmentation of acute ischemic lesions on DWI. The CNN algorithms were compared with conventional algorithms based on DWI and the apparent diffusion coefficient (ADC) signal intensity. The performances of the algorithms were assessed using the Dice index with 5-fold cross-validation. The Dice indices were analyzed according to infarct volumes (< 10 mL, ≥ 10 mL), number of infarcts (≤ 5, 6-10, ≥ 11), and b-value of 1000 (b1000) signal intensities (< 50, 50-100, > 100), time intervals to DWI, and DWI protocols. Results: The CNN algorithms were significantly superior to conventional algorithms (p < 0.001). Dice indices for the CNN algorithms were 0.85 for U-Net and DenseNet and 0.86 for an ensemble of U-Net and DenseNet, while the indices were 0.58 for ADC-b1000 and b1000-ADC and 0.52 for the commercial ADC algorithm. The Dice indices for small and large lesions, respectively, were 0.81 and 0.88 with U-Net, 0.80 and 0.88 with DenseNet, and 0.82 and 0.89 with the ensemble of U-Net and DenseNet. The CNN algorithms showed significant differences in Dice indices according to infarct volumes (p < 0.001). Conclusion: The CNN algorithm for automatic segmentation of acute ischemic lesions on DWI achieved Dice indices greater than or equal to 0.85 and showed superior performance to conventional algorithms.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.35 no.1
• /
• pp.19-24
• /
• 1999

This paper is second part on the auditory thresholds and fish behaviors to the underwater sounds for luring of target species at the set-net in the coast of Cheju. In order to obtain the critical ratio of yellow tails(Seriola quinqueradiata) and the emission level of underwater sound for luring of them, we make experiments to measure the auditory threshold of them using conditioning with electric shock. In state that the white noise with 10dB higher sound pressure level than ambient noise is emitted, the auditory thresholds of yellow tails are measured with 100~116.5dB and they are higher than those in state of no emission of white noise by the masking effects of it. Although sound pressure level of background noise go down, the auditory thresholds go up with frequency above than 300Hz.The critical ratio of yellow-tails in frequency of 80Hz, 100Hz, 200Hz, 500Hz, 800Hz are 46dB, 40dB, 50dB, 52dB, 60dB, 70dB respectively. The sound pressure level of which the signal sound is recognized by yellow tails under the ambient noise is above 100dB and the critical ratio of them is above 40dB.

• Proceedings of the Korean Geotechical Society Conference
• /
• 2007.09a
• /
• pp.558-569
• /
• 2007

It is important to obtain real-time data from long-term monitoring of landfills and develop leachate leakage detection system for the integrated management of landfills. A novel real time monitoring system and early leakage detection system was suggested in this study. The suggested monitoring system is composed of two parts; (1) a set of moisture sensors which monitor the areas surrounding the landfill, and (2) a set of moisture and temperature sensors which monitor the landfill inside. For the assessment for landfills stabilization, real-time monitoring system was evaluated in dry and wet cell of pilot-site. In addition, the grid-net electrical conductivity measurement system was also suggested as early leakage detection system. In this study, the field applicability of suggested systems was evaluated through pilot-scale field tests. The results of pilot-scale field model tests indicate that the grid-net electrical conductivity measurement method can be applicable to the detection of landfill leachate at the initial stage of intrusion, and thus has a potential for monitoring leachate leakage at waste landfills.