• Investigative Magnetic Resonance Imaging
• /
• v.25 no.3
• /
• pp.156-163
• /
• 2021

Recently, neuromelanin and nigrosome imaging techniques have been developed to evaluate the substantia nigra in Parkinson's disease. Previous studies have shown potential benefits of quantitative analysis of neuromelanin and nigrosome images in the substantia nigra, although visual assessments have been performed to evaluate structures in most studies. In this study, we investigate the potential of using deep learning based automatic region segmentation techniques for quantitative analysis of the substantia nigra. The deep convolutional neural network was trained to automatically segment substantia nigra regions on 3D nigrosome and neuromelanin sensitive MR images obtained from 30 subjects. With a 5-fold cross-validation, the mean calculated dice similarity coefficient between manual and deep learning was 0.70 ± 0.11. Although calculated dice similarity coefficients were relatively low due to empirically drawn margins, selected slices were overlapped for more than two slices of all subjects. Our results demonstrate that deep convolutional neural network-based method could provide reliable localization of substantia nigra regions on neuromelanin and nigrosome sensitive MR images.

• International Journal of Computer Science & Network Security
• /
• v.21 no.9
• /
• pp.51-62
• /
• 2021

Plant disease is one of the issues that can create losses in the production and economy of the agricultural sector. Early detection of this disease for finding solutions and treatments is still a challenge in the sustainable agriculture field. Currently, image processing techniques and machine learning methods have been applied to detect plant diseases successfully. However, the effectiveness of these methods still needs to be improved, especially in multiclass plant diseases classification. In this paper, a convolutional neural network with a batch normalization-based deep learning approach for classifying plant diseases is used to develop an automatic diagnostic assistance system for leaf diseases. The significance of using deep learning technology is to make the system be end-to-end, automatic, accurate, less expensive, and more convenient to detect plant diseases from their leaves. For evaluating the proposed model, an experiment is conducted on a public dataset contains 20654 images with 15 plant diseases. The experimental validation results on 20% of the dataset showed that the model is able to classify the 15 plant diseases labels with 96.4% testing accuracy and 0.168 testing loss. These results confirmed the applicability and effectiveness of the proposed model for the plant disease detection task.

• Nuclear Engineering and Technology
• /
• v.55 no.1
• /
• pp.100-108
• /
• 2023

In this study, a neural network model inspired by a one-dimensional convolution U-net is developed to automatically accelerate edge localized mode (ELM) detection from big diagnostic data of fusion devices and increase the detection accuracy regardless of the hyperparameter setting. This model recognizes the input signal patterns and overcomes the problems of existing detection algorithms, such as the prominence algorithm and those of differential methods with high sensitivity for the threshold and signal intensity. To train the model, 10 sets of discharge radiation data from the KSTAR are used and sliced into 11091 inputs of length 12 ms, of which 20% are used for validation. According to the receiver operating characteristic curves, our model shows a positive prediction rate and a true prediction rate of approximately 90% each, which is comparable to the best detection performance afforded by other algorithms using their optimized hyperparameters. The accurate and automatic ELM-burst detection methodology used in our model can be beneficial for determining plasma properties, such as the ELM frequency from big data measured in multiple experiments using machines from the KSTAR device and ITER. Additionally, it is applicable to feature detection in the time-series data of other engineering fields.

• Korean Journal of Radiology
• /
• v.21 no.1
• /
• pp.88-100
• /
• 2020

Objective: We aimed to develop and validate a deep learning system for fully automated segmentation of abdominal muscle and fat areas on computed tomography (CT) images. Materials and Methods: A fully convolutional network-based segmentation system was developed using a training dataset of 883 CT scans from 467 subjects. Axial CT images obtained at the inferior endplate level of the 3rd lumbar vertebra were used for the analysis. Manually drawn segmentation maps of the skeletal muscle, visceral fat, and subcutaneous fat were created to serve as ground truth data. The performance of the fully convolutional network-based segmentation system was evaluated using the Dice similarity coefficient and cross-sectional area error, for both a separate internal validation dataset (426 CT scans from 308 subjects) and an external validation dataset (171 CT scans from 171 subjects from two outside hospitals). Results: The mean Dice similarity coefficients for muscle, subcutaneous fat, and visceral fat were high for both the internal (0.96, 0.97, and 0.97, respectively) and external (0.97, 0.97, and 0.97, respectively) validation datasets, while the mean cross-sectional area errors for muscle, subcutaneous fat, and visceral fat were low for both internal (2.1%, 3.8%, and 1.8%, respectively) and external (2.7%, 4.6%, and 2.3%, respectively) validation datasets. Conclusion: The fully convolutional network-based segmentation system exhibited high performance and accuracy in the automatic segmentation of abdominal muscle and fat on CT images.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.9 no.1
• /
• pp.46-55
• /
• 2006

The purpose of this study is a development of an automatic mosaicing for applying to large number of airborne multispectral images, which reduces manual operation by human. 2436 airborne multispectral images were acquired from DuncanTech MS4100 camera with three bands; green, red and near infrared. LIDAR(LIght Detection And Ranging) data and GPS/INS(global positioning system/inertial navigation system) data were collected with the multispectral images. First, the multispectral images were converted to image patterns by unsupervised classification. Their patterns were compared with those of adjacent images to derive relative spatial position between images. Relative spatial positions were derived for 80% of the whole images. Second, it accomplished an automatic mosaicing using GPS/INS data and unsupervised classification. Since the time of GPS/INS data did not synchronized the time of readout images, synchronized GPS/INS data with the time of readout image were selected in consecutive data by comparing unsupervised classified images. This method realized mosaicing automatically for 96% images and RMSE (root mean square error) for the spatial precision of mosaiced images was only 1.44 m by validation with LIDAR data.

• The Journal of the Korea Contents Association
• /
• v.5 no.4
• /
• pp.169-179
• /
• 2005

Standard industry and occupation code are usually assigned manually in Korean census. The manual coding is very labor intensive and expensive task. Furthermore, inconsistent coding is resulted from the ability of human experts and their working environments. This paper proposes an automatic code classification system which converts natural language responses on survey questionnaires into corresponding numeric codes by using manually constructed rule base and example-based machine learning. The system was trained with 400,000 records of which standard codes was assigned. It was evaluated with 10-fold cross validation and was tested with three code sets: population occupation set, industry set, and industry survey set. The proposed system showed 76.63%, 82.24 and 99.68% accuracy for each code set.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.23 no.1
• /
• pp.19-30
• /
• 2005

Korean government has constructed a nationwide cadastral map database through the cadastral map computerization project and also produced a variety of spatial data through the NGIS (National Geographic Information Systems) project. Under this circumstance, it is needed to set up the new automatic methodology that effectively solve cadastral non-coincidence problems by using various digital map data instead of expensive field survey methods. This study proposed a new automatic methodology for cadastral non-coincidence surveying and developed a prototype system as a proof of concept. Validation of this proposed method was done with some test areas. Results showed that this methodology could easily detect and assess both regional non-coincidence levels and cadastral map quadrangle non-coincidence levels. We expect that this new methodology can provide many benefits in planning and determining work priority of the forthcoming nationwide cadastral re-surveying project.

• Journal of Korea Multimedia Society
• /
• v.13 no.10
• /
• pp.1565-1574
• /
• 2010

While we are driving a car, we are able to listen to musics by two ways: by selecting (manipulating) what we want and by just playing as they are given (in CD). These methods make a driver tired while he is driving or it means that a music which is provided is not concerned with a driver's preference. To improve these problems, there have been many studies about the automatic music providing systems based on driver's emotion. However, these studies have some difficult problems: the first one is that it is not easy to determine driver's emotion, and the other one is that it is hard to recommend and play the suitable music corresponding to the determined user's emotion. In this paper, to overcome the second problem mentioned above, a new mapping method between driver's emotion and music genre for automatic music providing system on vehicle is presented and two experiments are examined for the validation of the proposed method. The experimental results and discussions are explored to show the effectiveness and validity of the proposed method.

• Spatial Information Research
• /
• v.16 no.2
• /
• pp.193-206
• /
• 2008

If stereo images is used for Digital Elevation Model (DEM) generation, a DEM is generally made by matching left image against right image from stereo images. In stereo matching, tie-points are used as initial match candidate points. The number and distribution of tie-points influence the matching result. DEM made from matching result has errors such as holes, peaks, etc. These errors are usually interpolated by neighbored pixel values. In this paper, we propose the DEM generation method combined with automatic tie-points extraction using existing DEM, image pyramid, and interpolating new DEM using existing DEM for more reliable DEM. For test, we used IKONOS, QuickBird, SPOT5 stereo images and a DTED level 2 data. The test results show that the proposed method automatically makes reliable DEMs. For DEM validation, we compared heights of DEM by proposed method with height of existing DTED level 2 data. In comparison result, RMSE was under than 15 m.

• KIPS Transactions on Software and Data Engineering
• /
• v.5 no.9
• /
• pp.419-424
• /
• 2016

Named entity recognition is required to improve the retrieval accuracy of patent documents or similar patents in the claims and patent descriptions. In this paper, we proposed an automatic named entity recognition for patents by using a conditional random field that is one of the best methods in machine learning research. Named entity recognition system has been constructed from the training set of tagged corpus with 660,000 words and 70,000 words are used as a test set for evaluation. The experiment shows that the accuracy is 93.6% and the Kappa coefficient is 0.67 between manual tagging and automatic tagging system. This figure is better than the Kappa coefficient 0.6 for manually tagged results and it shows that automatic named entity tagging system can be used as a practical tagging for patent documents in replacement of a manual tagging.