• Journal of Korea Multimedia Society
• /
• v.20 no.2
• /
• pp.205-215
• /
• 2017

The brain magnetic resonance images (MRI) is an important imaging biomarker in Alzheimer's disease (AD) as the cerebral atrophy has been shown to strongly associate with cognitive symptoms. The decrease of volume estimates in different structures of the medial temporal lobe related to memory correlates with the decline of cognitive functions in neurodegenerative diseases. During the past decades several methods have been developed for quantifying the disease related atrophy of hippocampus from MRI. Special effort has been dedicated to separate AD and mild cognitive impairment (MCI) related modifications from normal aging for the purpose of early detection and prediction. We trained a multi-class support vector machine (SVM) with probabilistic outputs on a sample (n = 58) of 20 normal controls (NC), 19 individuals with MCI, and 19 individuals with AD. The model was then applied to the cross-validation of same data set which no labels were known and the predictions. This study presents data on the association between MRI quantitative parameters of hippocampus and its quantitative structural changes examination use on the classification of the diseases.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.55 no.3
• /
• pp.131-137
• /
• 2006

The secondary system of nuclear power plants consists of sophisticated piping systems operating in very aggressive erosion and corrosion environments, which make a piping system vulnerable to the wear and degradation due to the several chemical components and high flow rate (~10 m/sec) of the coolant. To monitor the wear and degradation on a pipe, the vibration signals are measured from the pipe with an accelerometer For analyzing the vibration signal the time-frequency analysis (TFA) is used, which is known to be effective for the analysis of time-varying or transient signals. To reduce the inteferences (cross-terms) due to the bilinear structure of the time-frequency distribution, an adaptive cone-kernel distribution (ACKD) is proposed. The cone length of ACKD to determine the characteristics of distribution is optimally selected through an adaptive algorithm using the normalized Shannon's entropy And the ACKD's are compared with the results of other analyses based on the Fourier Transform (FT) and other TFA's. The ACKD shows a better signature for the wear/degradation within a pipe and provides the additional information in relation to the time that any analysis based on the conventional FT can not provide.

• Bulletin of the Korean Chemical Society
• /
• v.21 no.1
• /
• pp.110-117
• /
• 2000

A series of new liquid crystalline dimesogenic compounds with chiral tails was synthesized, and their thermal and liquid crystalline properties were studied. The chain length of the central polymethylene spacers (x) was varied from dimethylene (2) to decamethylene (12). These compounds were characterized by elemental analysis, IR and NMR spectroscopy, differential scanning calorimetry (DSC), and cross-polarizing microscopy. All compounds were found to be enantiotropically liquid crystalline, and the values of melting ($T_m$) and isotropization temperature ($T_i$) as well as enthalpy change (Δ$H_i$) and entropy change for isotropization (Δ$S_i$) decreased in a zig-zag fashion revealing the so-called odd-even effect as x increases. Their mesomorphic properties fall into three categories depending upon x; (a) compounds with x=2 and 4 formed two different mesophases, smectic and cholesteric phases in that order on heating, and vice versa on cooling, (b) compounds with x=3, 7, 8, 10 and 11 reversibly formed only the cholesteric phase, and (c) compounds with x=5, 6, 9 and 12 exhibited only a cholesteric phase on heating, whereas on cooling they formed two different mesophases, cholesteric and smectic phases, sequentially.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.49 no.10
• /
• pp.883-892
• /
• 2021

This paper provides a process for developing deep learning-based aerial object detection models that can run in realtime on onboard. To improve object detection performance, we pre-process and augment the training data in the training stage. In addition, we perform transfer learning and apply a weighted cross-entropy method to reduce the variations of detection performance for each class. To improve the inference speed, we have generated inference acceleration engines with quantization. Then, we analyze the real-time performance and detection performance on custom aerial image dataset to verify generalization.

• Journal of Korea Multimedia Society
• /
• v.21 no.12
• /
• pp.1425-1430
• /
• 2018

In this paper, we propose a CNN based deep learning algorithm for semantic segmentation of images. In order to improve the accuracy of semantic segmentation, we combined pixel level object classification and image level object classification. The image level object classification is used to accurately detect the characteristics of an image, and the pixel level object classification is used to indicate which object area is included in each pixel. The proposed network structure consists of three parts in total. A part for extracting the features of the image, a part for outputting the final result in the resolution size of the original image, and a part for performing the image level object classification. Loss functions exist for image level and pixel level classification, respectively. Image-level object classification uses KL-Divergence and pixel level object classification uses cross-entropy. In addition, it combines the layer of the resolution of the network extracting the features and the network of the resolution to secure the position information of the lost feature and the information of the boundary of the object due to the pooling operation.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.9
• /
• pp.4665-4683
• /
• 2019

Previous methods build image annotation model by leveraging three basic dependencies: relations between image and label (image/label), between images (image/image) and between labels (label/label). Even though plenty of researches show that multiple dependencies can work jointly to improve annotation performance, different dependencies actually do not "work jointly" in their diagram, whose performance is largely depending on the result predicted by image/label section. To address this problem, we propose the adaptive attention annotation model (AAAM) to associate these dependencies with the prediction path, which is composed of a series of labels (tags) in the order they are detected. In particular, we optimize the prediction path by detecting the relevant labels from the easy-to-detect to the hard-to-detect, which are found using Binary Cross-Entropy (BCE) and Triplet Margin (TM) losses, respectively. Besides, in order to capture the inforamtion of each label, instead of explicitly extracting regional featutres, we propose the self-attention machanism to implicitly enhance the relevant region and restrain those irrelevant. To validate the effective of the model, we conduct experiments on three well-known public datasets, COCO 2014, IAPR TC-12 and NUSWIDE, and achieve better performance than the state-of-the-art methods.

• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.153-166
• /
• 2022

Identifying fine cracks in steel bridge facilities is a challenging task of structural health monitoring (SHM). This study proposed an end-to-end crack image segmentation framework based on a one-step Convolutional Neural Network (CNN) for pixel-level object recognition with high accuracy. To particularly address the challenges arising from small object detection in complex background, efforts were made in loss function selection aiming at sample imbalance and module modification in order to improve the generalization ability on complicated images. Specifically, loss functions were compared among alternatives including the Binary Cross Entropy (BCE), Focal, Tversky and Dice loss, with the last three specialized for biased sample distribution. Structural modifications with dilated convolution, Spatial Pyramid Pooling (SPP) and Feature Pyramid Network (FPN) were also performed to form a new backbone termed CrackDet. Models of various loss functions and feature extraction modules were trained on crack images and tested on full-scale images collected on steel box girders. The CNN model incorporated the classic U-Net as its backbone, and Dice loss as its loss function achieved the highest mean Intersection-over-Union (mIoU) of 0.7571 on full-scale pictures. In contrast, the best performance on cropped crack images was achieved by integrating CrackDet with Dice loss at a mIoU of 0.7670.

• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.221-235
• /
• 2022

Manual inspection of steel box girders on long span bridges is time-consuming and labor-intensive. The quality of inspection relies on the subjective judgements of the inspectors. This study proposes an automated approach to detect and segment cracks in high-resolution images. An end-to-end cascaded framework is proposed to first detect the existence of cracks using a deep convolutional neural network (CNN) and then segment the crack using a modified U-Net encoder-decoder architecture. A Naïve Bayes data fusion scheme is proposed to reduce the false positives and false negatives effectively. To generate the binary crack mask, first, the original images are divided into 448 × 448 overlapping image patches where these image patches are classified as cracks versus non-cracks using a deep CNN. Next, a modified U-Net is trained from scratch using only the crack patches for segmentation. A customized loss function that consists of binary cross entropy loss and the Dice loss is introduced to enhance the segmentation performance. Additionally, a Naïve Bayes fusion strategy is employed to integrate the crack score maps from different overlapping crack patches and to decide whether a pixel is crack or not. Comprehensive experiments have demonstrated that the proposed approach achieves an 81.71% mean intersection over union (mIoU) score across 5 different training/test splits, which is 7.29% higher than the baseline reference implemented with the original U-Net.

• Smart Media Journal
• /
• v.13 no.8
• /
• pp.16-25
• /
• 2024

Recently, there has been a growing emphasis on identifying both known and unknown diseases in plant disease recognition. In this task, a model trained only on images of known classes is required to classify an input image into either one of the known classes or into an unknown class. Consequently, the capability to recognize unknown diseases is critical for model deployment. To enhance this capability, we are considering three factors. Firstly, we propose a new logits-based scoring function for unknown scores. Secondly, initial experiments indicate that a compact feature space is crucial for the effectiveness of logits-based methods, leading us to employ the AM-Softmax loss instead of Cross-entropy loss during training. Thirdly, drawing inspiration from the efficacy of transfer learning, we utilize a large plant-relevant dataset, PlantCLEF2022, for pre-training a model. The experimental results suggest that our method outperforms current algorithms. Specifically, our method achieved a performance of 97.90 CSA, 91.77 AUROC, and 90.63 OSCR with the ResNet50 model and a performance of 98.28 CSA, 92.05 AUROC, and 91.12 OSCR with the ConvNext base model. We believe that our study will contribute to the community.

• Smart Structures and Systems
• /
• v.33 no.4
• /
• pp.253-262
• /
• 2024

Current-based mechanical fault diagnosis is more convenient and low cost since additional sensors are not required. However, it is still challenging to achieve this goal due to the weak fault information in current signals. In this paper, a dual-loss convolutional neural network (DLCNN) is proposed to implement the intelligent bearing fault diagnosis via current signals. First, a novel similarity loss (SimL) function is developed, which is expected to maximize the intra-class similarity and minimize the inter-class similarity in the model optimization operation. In the loss function, a weight parameter is further introduced to achieve a balance and leverage the performance of SimL function. Second, the DLCNN model is constructed using the presented SimL and the cross-entropy loss. Finally, the two-phase current signals are fused and then fed into the DLCNN to provide more fault information. The proposed DLCNN is tested by experiment data, and the results confirm that the DLCNN achieves higher accuracy compared to the conventional CNN. Meanwhile, the feature visualization presents that the samples of different classes are separated well.