• Journal of Biomedical Engineering Research
• /
• v.41 no.2
• /
• pp.94-100
• /
• 2020

The purpose of this study was to present the models for classifying the wrist X-ray images by types and for segmenting the radius automatically in each image using deep learning and to verify the learned models. The data were a total of 904 wrist X-rays with the distal radius fracture, consisting of 472 anteroposterior (AP) and 432 lateral images. The learning model was the ResNet50 model for AP/lateral image classification, and the U-Net model for segmentation of the radius. In the model for AP/lateral image classification, 100.0% was showed in precision, recall, and F1 score and area under curve (AUC) was 1.0. The model for segmentation of the radius showed an accuracy of 99.46%, a sensitivity of 89.68%, a specificity of 99.72%, and a Dice similarity coefficient of 90.05% in AP images and an accuracy of 99.37%, a sensitivity of 88.65%, a specificity of 99.69%, and a Dice similarity coefficient of 86.05% in lateral images. The model for AP/lateral classification and the segmentation model of the radius learned through deep learning showed favorable performances to expect clinical application.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.11
• /
• pp.1615-1622
• /
• 2022

In order to overcome the limitations of the instructor-centered lecture-style teaching method, recently, flipped learning, a learner-centered teaching method, has been widely introduced. However, despite the many advantages of flipped learning, there is a problem that students cannot solve questions that arise during prior learning in real time. Therefore, in order to solve this problem, we developed DBbot, an assistant chatbot for database design course managed in the flipped learning method. The DBBot is composed of a chatbot app for learners and a chatbot management app for instructors. Also, it's implemented so that questions that instructors can anticipate in advance, such as questions related to class operation and every semester repeated questions related to learning content, can be answered using Google's DialogFlow. It's implemented so that questions that the instructor cannot predict in advance, such as questions related to team projects, can be answered using the question/answer DB and the BM25 algorithm, which is a similarity comparison algorithm.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.2
• /
• pp.19-25
• /
• 2023

In this paper, we propose a method for improving raw images captured in a low light condition based on deep learning considering the image signal processing. In the case of a smart phone camera, compared to a DSLR camera, the size of a lens or sensor is limited, so the noise increases and the reduces the quality of images in low light conditions. Existing deep learning-based low-light image processing methods create unnatural images in some cases since they do not consider the lens shading effect and white balance, which are major factors in the image signal processing. In this paper, pixel distances from the image center and channel average values are used to consider the lens shading effect and white balance with a deep learning model. Experiments with low-light images taken with a smart phone demonstrate that the proposed method achieves a higher peak signal to noise ratio and structural similarity index measure than the existing method by creating high-quality low-light images.

• Journal of IKEEE
• /
• v.24 no.2
• /
• pp.514-522
• /
• 2020

In Continuous Learning environment, we identify that the Catastrophic Forgetting phenomenon, which forgets the information of previously learned data, occurs easily between data having different domains. To control this phenomenon, we introduce how to measure the relationship between previously learned data and newly learned data through the distribution of the neural network's output, and how to use these measurements to mitigate the Catastrophic Forcing phenomenon. MNIST and EMNIST data were used for evaluation, and experiments showed an average 22.37% improvement in accuracy for previous data.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.31B no.9
• /
• pp.133-141
• /
• 1994

Conventional competitive learning algorithms compute the Euclidien distance to determine the winner neuron out of all predetermined output neurons. In such cases, there is a drawback that the performence of the learning algorithm depends on the initial reference(=weight) vectors. In this paper, we propose a new competitive learning algorithm that dynamically generates output neurons. The proposed method generates output neurons by dynamically changing the class thresholds for all output neurons. We compute the similarity between the input vector and the reference vector of each output neuron generated. If the two are similar, the reference vector is adjusted to make it still more like the input vector. Otherwise, the input vector is designated as the reference vector of a new outputneuron. Since the reference vectors of output neurons are dynamically assigned according to input pattern distribution, the proposed method gets around the phenomenon that learning is early determined due to redundant output neurons. Experiments using speech data have shown the proposed method to be superior to existint methods.

• Journal of KIISE:Software and Applications
• /
• v.37 no.4
• /
• pp.297-304
• /
• 2010

Clustering method which based on sentence type or document genre is a technique used to improve translation quality of SMT(statistical machine translation) by domain-specific translation. But there is no previous research using sentence type and document genre information simultaneously. In this paper, we suggest an integrated clustering method that classifying sentence type by syntactic structure similarity and document genre by word similarity information. We interpolated domain-specific models from clusters with general models to improve translation quality of SMT system. Kernel function and cosine measures are applied to calculate structural similarity and word similarity. With these similarities, we used machine learning algorithms similar to K-means to clustering. In Japanese-English patent translation corpus, we got 2.5% point relative improvements of translation quality at optimal case.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09a
• /
• pp.419-422
• /
• 2003

To improve the recognition accuracy of a developed artificial odor discrimination system for three mixture fragrance recognition, Fuzzy Similarity based Self-Organized Network inspired by Immune Algorithm (F-SONIA) is proposed. Minimum, average, and maximum values of fragrance data acquisitions are used to form triangular fuzzy numbers. Then the fuzzy similarity treasure is used to define the relationship between fragrance inputs and connection strengths of hidden units. The fuzzy similarity is defined as the maximum value of the intersection region between triangular fuzzy set of input vectors and the connection strengths of hidden units. In experiments, performances of the proposed method is compared with the conventional Self-Organized Network inspired by Immune Algorithm (SONIA), and the Fuzzy Learning Vector Quantization (FLVQ). Experiments show that F-SONIA improves recognition accuracy of SONIA by 3-9％. Comparing to the previously developed artificial odor discrimination system that used FLVQ as pattern classifier, the recognition accuracy is increased by 14-25％.

• Journal of Institute of Control, Robotics and Systems
• /
• v.5 no.5
• /
• pp.550-557
• /
• 1999

This paper describes a road-following controller using the proposed neural network for autonomous vehicle. Road-following with visual sensor like camera requires intelligent control algorithm because analysis of relation from road image to steering control is complex. The proposed neural network, relative similarity modular network(RSMN), is composed of some learning networks and a partitioniing network. The partitioning network divides input space into multiple sections by similarity of input data. Because divided section has simlar input patterns, RSMN can learn nonlinear relation such as road-following with visual control easily. Visual control uses two criteria on road image from camera; one is position of vanishing point of road, the other is slope of vanishing line of road. The controller using neural network has input of two criteria and output of steering angle. To confirm performance of the proposed neural network controller, a software is developed to simulate vehicle dynamics, camera image generation, visual control, and road-following. Also, prototype autonomous electric vehicle is developed, and usefulness of the controller is verified by physical driving test.

• KIPS Transactions on Software and Data Engineering
• /
• v.7 no.12
• /
• pp.469-476
• /
• 2018

In order to jointly utilize research infrastructure and to build efficient construction, which are essential in science and technology research and development process. Although various classification methods have been introduced for efficient utilization of registered information, functions that can be directly utilized such as similar research infrastructure search is not yet been implemented due to limitations of collection information. In this study, we analyzed the similar search technique so far, presented the methodology for the calculation of similarity of research infrastructure, and analyzed the learning result. Study suggested that a technique can be use to extract meaningful keywords from information and analyze the similarity between the research infrastructure.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.1
• /
• pp.545-550
• /
• 2022

In Deep Learning method, it is well known that it requires large amount of data to train the deep neural network. And it also requires the labeling of each data to fully train the neural network, which means that experts should spend lots of time to provide the labeling. To alleviate the problem of time-consuming labeling process, some methods have been suggested such as weak-supervised method, one-shot learning, self-supervised, suggestive learning, and so on. In this manuscript, those methods are analyzed and its possible future direction of the research is suggested.