• Korean Journal of Remote Sensing
• /
• v.35 no.5_1
• /
• pp.681-692
• /
• 2019

In this paper, a hybrid deep learning model, called 2D convolution with bidirectional long short-term memory (2DCBLSTM), is presented that can effectively combine both spatial and temporal features for crop classification. In the proposed model, 2D convolution operators are first applied to extract spatial features of crops and the extracted spatial features are then used as inputs for a bidirectional LSTM model that can effectively process temporal features. To evaluate the classification performance of the proposed model, a case study of crop classification was carried out using multi-temporal unmanned aerial vehicle images acquired in Anbandegi, Korea. For comparison purposes, we applied conventional deep learning models including two-dimensional convolutional neural network (CNN) using spatial features, LSTM using temporal features, and three-dimensional CNN using spatio-temporal features. Through the impact analysis of hyper-parameters on the classification performance, the use of both spatial and temporal features greatly reduced misclassification patterns of crops and the proposed hybrid model showed the best classification accuracy, compared to the conventional deep learning models that considered either spatial features or temporal features. Therefore, it is expected that the proposed model can be effectively applied to crop classification owing to its ability to consider spatio-temporal features of crops.

• Journal of the Korea Computer Graphics Society
• /
• v.24 no.5
• /
• pp.41-48
• /
• 2018

The most natural way to increase immersion and provide free interaction in a virtual environment is to provide a gesture interface using the user's hand. However, most studies about hand gesture recognition require specialized sensors or equipment, or show low recognition rates. This paper proposes a three-dimensional DenseNet Convolutional Neural Network that enables recognition of hand gestures with no sensors or equipment other than an RGB camera for hand gesture input and introduces a virtual reality game based on it. Experimental results on 4 static hand gestures and 6 dynamic hand gestures showed that they could be used as real-time user interfaces for virtual reality games with an average recognition rate of 94.2% at 50ms. Results of this research can be used as a hand gesture interface not only for games but also for education, medicine, and shopping.

• The Journal of the Acoustical Society of Korea
• /
• v.42 no.3
• /
• pp.214-220
• /
• 2023

In this paper, we propose a method to improve texture recognition performance from haptic acceleration signals representing the texture characteristics of object surface materials by using a Conformer model that combines the advantages of a convolutional neural network and a transformer. In the proposed method, three-axis acceleration signals generated by impact sound and vibration are combined into one-dimensional acceleration data while a person contacts the surface of the object materials using a tool such as a stylus , and the logarithmic Mel-spectrogram is extracted from the haptic acceleration signal similar to the audio signal. Then, Conformer is applied to the extracted the logarithmic Mel-spectrogram to learn main local and global frequency features in recognizing the texture of various object materials. Experiments on the Lehrstuhl für Medientechnik (LMT) haptic texture dataset consisting of 60 materials to evaluate the performance of the proposed model showed that the proposed method can effectively recognize the texture of the object surface material better than the existing methods.

• KIPS Transactions on Software and Data Engineering
• /
• v.7 no.3
• /
• pp.91-98
• /
• 2018

Automatic detection of pig wasting diseases is an important issue in the management of group-housed pigs. In particular, porcine respiratory diseases are one of the main causes of mortality among pigs and loss of productivity in intensive pig farming. In this paper, we propose a noise-robust system for the early detection and recognition of pig wasting diseases using sound data. In this method, first we convert one-dimensional sound signals to two-dimensional gray-level images by normalization, and extract texture images by means of dominant neighborhood structure technique. Lastly, the texture features are then used as inputs of convolutional neural networks as an early anomaly detector and a respiratory disease classifier. Our experimental results show that this new method can be used to detect pig wasting diseases both economically (low-cost sound sensor) and accurately (over 96% accuracy) even under noise-environmental conditions, either as a standalone solution or to complement known methods to obtain a more accurate solution.

• Smart Structures and Systems
• /
• v.29 no.4
• /
• pp.625-640
• /
• 2022

Maglev rail joints are vital components serving as connections between the adjacent F-type rail sections in maglev guideway. Damage to maglev rail joints such as bolt looseness may result in rough suspension gap fluctuation, failure of suspension control, and even sudden clash between the electromagnets and F-type rail. The condition monitoring of maglev rail joints is therefore highly desirable to maintain safe operation of maglev. In this connection, an online damage detection approach based on three-dimensional (3D) convolutional neural network (CNN) and time-frequency characterization is developed for simultaneous detection of multiple damage of maglev rail joints in this paper. The training and testing data used for condition evaluation of maglev rail joints consist of two months of acceleration recordings, which were acquired in-situ from different rail joints by an integrated online monitoring system during a maglev train running on a test line. Short-time Fourier transform (STFT) method is applied to transform the raw monitoring data into time-frequency spectrograms (TFS). Three CNN architectures, i.e., small-sized CNN (S-CNN), middle-sized CNN (M-CNN), and large-sized CNN (L-CNN), are configured for trial calculation and the M-CNN model with excellent prediction accuracy and high computational efficiency is finally optioned for multiple damage detection of maglev rail joints. Results show that the rail joints in three different conditions (bolt-looseness-caused rail step, misalignment-caused lateral dislocation, and normal condition) are successfully identified by the proposed approach, even when using data collected from rail joints from which no data were used in the CNN training. The capability of the proposed method is further examined by using the data collected after the loosed bolts have been replaced. In addition, by comparison with the results of CNN using frequency spectrum and traditional neural network using TFS, the proposed TFS-CNN framework is proven more accurate and robust for multiple damage detection of maglev rail joints.

• Current Optics and Photonics
• /
• v.8 no.2
• /
• pp.127-137
• /
• 2024

The distribution of tags is an important factor that affects the performance of radio-frequency identification (RFID). To study RFID performance, it is necessary to obtain RFID tags' coordinates. However, the positioning method of RFID technology has large errors, and is easily affected by the environment. Therefore, a new method using optical measurement is proposed to achieve RFID performance analysis. First, due to the possibility of blurring during image acquisition, the paper derives a new image prior to removing blurring. A nonlocal means-based method for image deconvolution is proposed. Experimental results show that the PSNR and SSIM indicators of our algorithm are better than those of a learning deep convolutional neural network and fast total variation. Second, an RFID dynamic testing system based on photoelectric sensing technology is designed. The reading distance of RFID and the three-dimensional coordinates of the tags are obtained. Finally, deep learning is used to model the RFID reading distance and tag distribution. The error is 3.02%, which is better than other algorithms such as a particle-swarm optimization back-propagation neural network, an extreme learning machine, and a deep neural network. The paper proposes the use of optical methods to measure and collect RFID data, and to analyze and predict RFID performance. This provides a new method for testing RFID performance.

• Korean Journal of Radiology
• /
• v.22 no.2
• /
• pp.168-178
• /
• 2021

Objective: To provide an automatic method for segmentation and diameter measurement of type B aortic dissection (TBAD). Materials and Methods: Aortic computed tomography angiographic images from 139 patients with TBAD were consecutively collected. We implemented a deep learning method based on a three-dimensional (3D) deep convolutional neural (CNN) network, which realizes automatic segmentation and measurement of the entire aorta (EA), true lumen (TL), and false lumen (FL). The accuracy, stability, and measurement time were compared between deep learning and manual methods. The intra- and inter-observer reproducibility of the manual method was also evaluated. Results: The mean dice coefficient scores were 0.958, 0.961, and 0.932 for EA, TL, and FL, respectively. There was a linear relationship between the reference standard and measurement by the manual and deep learning method (r = 0.964 and 0.991, respectively). The average measurement error of the deep learning method was less than that of the manual method (EA, 1.64% vs. 4.13%; TL, 2.46% vs. 11.67%; FL, 2.50% vs. 8.02%). Bland-Altman plots revealed that the deviations of the diameters between the deep learning method and the reference standard were -0.042 mm (-3.412 to 3.330 mm), -0.376 mm (-3.328 to 2.577 mm), and 0.026 mm (-3.040 to 3.092 mm) for EA, TL, and FL, respectively. For the manual method, the corresponding deviations were -0.166 mm (-1.419 to 1.086 mm), -0.050 mm (-0.970 to 1.070 mm), and -0.085 mm (-1.010 to 0.084 mm). Intra- and inter-observer differences were found in measurements with the manual method, but not with the deep learning method. The measurement time with the deep learning method was markedly shorter than with the manual method (21.7 ± 1.1 vs. 82.5 ± 16.1 minutes, p < 0.001). Conclusion: The performance of efficient segmentation and diameter measurement of TBADs based on the 3D deep CNN was both accurate and stable. This method is promising for evaluating aortic morphology automatically and alleviating the workload of radiologists in the near future.

• Journal of Broadcast Engineering
• /
• v.23 no.6
• /
• pp.886-895
• /
• 2018

In this paper, we introduce a technique to detect the video forgery by using the regularity that occurs in the video compression process. The proposed method uses the hierarchical regularity lost by the video double compression and the frame deletion. In order to extract such irregularities, the depth information of CU and TU, which are basic units of HEVC, is used. For improving performance, we make a depth map of CU and TU using local information, and then create input data by grouping them in GoP units. We made a decision whether or not the video is double-compressed and forged by using a general three-dimensional convolutional neural network. Experimental results show that it is more effective to detect whether or not the video is forged compared with the results using the existing machine learning algorithm.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.5
• /
• pp.1967-1983
• /
• 2020

The goal of this paper is to analyze and build an algorithm to recognize hand gestures applying to smart home applications. The proposed algorithm uses image processing techniques combing with artificial neural network (ANN) approaches to help users interact with computers by common gestures. We use five types of gestures, namely those for Stop, Forward, Backward, Turn Left, and Turn Right. Users will control devices through a camera connected to computers. The algorithm will analyze gestures and take actions to perform appropriate action according to users requests via their gestures. The results show that the average accuracy of proposal algorithm is 92.6 percent for images and more than 91 percent for video, which both satisfy performance requirements for real-world application, specifically for smart home services. The processing time is approximately 0.098 second with 10 frames/sec datasets. However, accuracy rate still depends on the number of training images (video) and their resolution.

• Journal of the Korea Computer Graphics Society
• /
• v.30 no.3
• /
• pp.109-123
• /
• 2024

Research into vision-based end-to-end autonomous driving systems utilizing deep learning and reinforcement learning has been steadily increasing. These systems typically encode continuous and high-dimensional vehicle states, such as location, velocity, orientation, and sensor data, into latent features, which are then decoded into a vehicular control policy. The complexity of urban driving environments necessitates the use of state representation learning through networks like Variational Autoencoders (VAEs) or Convolutional Neural Networks (CNNs). This paper analyzes the impact of different image state encoding methods on reinforcement learning performance in autonomous driving. Experiments were conducted in the CARLA simulator using RGB images and semantically segmented images captured by the vehicle's front camera. These images were encoded using VAE and Vision Transformer (ViT) networks. The study examines how these networks influence the agents' learning outcomes and experimentally demonstrates the role of each state representation technique in enhancing the learning efficiency and decision- making capabilities of autonomous driving systems.