• Journal of the Korea Convergence Society
• /
• v.12 no.10
• /
• pp.27-35
• /
• 2021

Automatic post editing is a research field that aims to automatically correct errors in machine translation results. This research is mainly being focus on high resource language pairs, such as English-German. Recent APE studies are mainly adopting transfer learning based research, where pre-training language models, or translation models generated through self-supervised learning methodologies are utilized. While translation based APE model shows superior performance in recent researches, as such researches are conducted on the high resource languages, the same perspective cannot be directly applied to the low resource languages. In this work, we apply two transfer learning strategies to Korean-English APE studies and show that transfer learning with translation model can significantly improves APE performance.

• 유체기계공업학회:학술대회논문집
• /
• 1999.12a
• /
• pp.317-323
• /
• 1999

The necessity of diagnosis of the rotating machinery which is widely used in the industry is increasing. Many research has been conducted to manipulate field vibration signal data for diagnosing the fault of designated machinery. As the pattern recognition tool of that signal, neural network which use usually back-propagation algorithm was used in the diagnosis of rotating machinery. In this paper, self-organizing feature map(SOFM) which is unsupervised learning algorithm is used in the abnormal vibration diagnosis of rotating machinery and then learning vector quantization(LVQ) which is supervised teaming algorithm is used to improve the quality of the classifier decision regions.

• ETRI Journal
• /
• v.46 no.3
• /
• pp.513-525
• /
• 2024

Deep neural networks trained on labeled medical data face major challenges owing to the economic costs of data acquisition through expensive medical imaging devices, expert labor for data annotation, and large datasets to achieve optimal model performance. The heterogeneity of diseases, such as Alzheimer's disease, further complicates deep learning because the test cases may substantially differ from the training data, possibly increasing the rate of false positives. We propose a reconstruction-based self-supervised anomaly detection model to overcome these challenges. It has a dual-subnetwork encoder that enhances feature encoding augmented by skip connections to the decoder for improving the gradient flow. The novel encoder captures local and global features to improve image reconstruction. In addition, we introduce an entropy-based image conversion method. Extensive evaluations show that the proposed model outperforms benchmark models in anomaly detection and classification using an encoder. The supervised and unsupervised models show improved performances when trained with data preprocessed using the proposed image conversion method.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.1
• /
• pp.180-194
• /
• 2021

For small image registration, feature-based approaches are likely to fail as feature detectors cannot detect enough feature points from low-resolution images. The classic FFT approach's prediction accuracy is high, but the registration time can be relatively long, about several seconds to register one image pair. To achieve real-time and high-precision rigid registration for small images, we apply deep neural networks for supervised rigid transformation prediction, which directly predicts the transformation parameters. We train deep registration models with rigidly transformed CIFAR-10 images and STL-10 images, and evaluate the generalization ability of deep registration models with transformed CIFAR-10 images, STL-10 images, and randomly generated images. Experimental results show that the deep registration models we propose can achieve comparable accuracy to the classic FFT approach for small CIFAR-10 images (32×32) and our LSTM registration model takes less than 1ms to register one pair of images. For moderate size STL-10 images (96×96), FFT significantly outperforms deep registration models in terms of accuracy but is also considerably slower. Our results suggest that deep registration models have competitive advantages over conventional approaches, at least for small images.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.8
• /
• pp.2787-2800
• /
• 2022

The aging society increases emergency situations of the elderly living alone and a variety of social crimes. In order to prevent them, techniques to detect emergency situations through voice are actively researched. This study proposes CutPaste-based anomaly detection model using multi-scale feature extraction in time series streaming data. In the proposed method, an audio file is converted into a spectrogram. In this way, it is possible to use an algorithm for image data, such as CNN. After that, mutli-scale feature extraction is applied. Three images drawn from Adaptive Pooling layer that has different-sized kernels are merged. In consideration of various types of anomaly, including point anomaly, contextual anomaly, and collective anomaly, the limitations of a conventional anomaly model are improved. Finally, CutPaste-based anomaly detection is conducted. Since the model is trained through self-supervised learning, it is possible to detect a diversity of emergency situations as anomaly without labeling. Therefore, the proposed model overcomes the limitations of a conventional model that classifies only labelled emergency situations. Also, the proposed model is evaluated to have better performance than a conventional anomaly detection model.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.12
• /
• pp.519-524
• /
• 2023

In hashing-based image retrieval, the hash code of a manipulated image is different from the original image, making it difficult to search for the same image. This paper proposes and evaluates a self-supervised deephashing model that generates perceptual hash codes from feature information such as texture, shape, and color of images. The comparison models are autoencoder-based variational inference models, but the encoder is designed with a fully connected layer, convolutional neural network, and transformer modules. The proposed model is a variational inference model that includes a SimAM module of extracting geometric patterns and positional relationships within images. The SimAM module can learn latent vectors highlighting objects or local regions through an energy function using the activation values of neurons and surrounding neurons. The proposed method is a representation learning model that can generate low-dimensional latent vectors from high-dimensional input images, and the latent vectors are binarized into distinguishable hash code. From the experimental results on public datasets such as CIFAR-10, ImageNet, and NUS-WIDE, the proposed model is superior to the comparative model and analyzed to have equivalent performance to the supervised learning-based deephashing model. The proposed model can be used in application systems that require low-dimensional representation of images, such as image search or copyright image determination.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.3
• /
• pp.670-684
• /
• 2024

3D Cross-Modal Retrieval (3DCMR) is a task that retrieves 3D objects regardless of modalities, such as images, meshes, and point clouds. One of the most prominent methods used for 3DCMR is the Cross-Modal Center Loss Function (CLF) which applies the conventional center loss strategy for 3D cross-modal search and retrieval. Since CLF is based on center loss, the center features in CLF are also susceptible to subtle changes in hyperparameters and external inferences. For instance, performance degradation is observed when the batch size is too small. Furthermore, the Mean Squared Error (MSE) used in CLF is unable to adapt to changes in batch size and is vulnerable to data variations that occur during actual inference due to the use of simple Euclidean distance between multi-modal features. To address the problems that arise from small batch training, we propose a Noisy Center Loss (NCL) method to estimate the optimal center features. In addition, we apply the simple Siamese representation learning method (SimSiam) during optimal center feature estimation to compare projected features, making the proposed method robust to changes in batch size and variations in data. As a result, the proposed approach demonstrates improved performance in ModelNet40 dataset compared to the conventional methods.

• Broadcasting and Media Magazine
• /
• v.27 no.2
• /
• pp.19-25
• /
• 2022

본 고에서는 최근 컴퓨터 비전 분야에서 가장 활발히 연구되고 있는 분야 중에 하나인 자가 학습(Self-supervised Learning) 기술의 동향과 향후 방향성에 대해서 논의한다. 컴퓨터 비전 분야에서의 자가 학습 기술은 최근에 Contrastive Learning 기법을 활용하여 활발하게 연구되고 있는데, 이를 위한 좋은 Positive와 Negative를 어떻게 추출할까에 대한 고민으로 수많은 연구들이 진행되어 왔다. 본 고에서는 이러한 방향성에서 대표적인 몇 가지의 방법론에 대해서 논의하고 이의 한계점을 언급하며 컴퓨터 비전 분야에서 자가 학습 기법이 가야 할 방향성에 대해서 논의하고자 한다.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.31 no.4
• /
• pp.351-359
• /
• 2011

In a structure, damage can occur at several scales from micro-cracking to corrosion or loose bolts. This makes the identification of damage difficult with one mode of sensing. Hence, a multi-mode actuated sensing system is proposed based on a self-sensing circuit using a piezoelectric sensor. In the self sensing-based multi-mode actuated sensing, one mode provides a wide frequency-band structural response from the self-sensed impedance measurement and the other mode provides a specific frequency-induced structural wavelet response from the self-sensed guided wave measurement. In this study, an experimental study on the pipeline system is carried out to verify the effectiveness and the robustness of the proposed structural health monitoring approach. Different types of structural damage are artificially inflicted on the pipeline system. To classify the multiple types of structural damage, a supervised learning-based statistical pattern recognition is implemented by composing a two-dimensional space using the damage indices extracted from the impedance and guided wave features. For more systematic damage classification, several control parameters to determine an optimal decision boundary for the supervised learning-based pattern recognition are optimized. Finally, further research issues will be discussed for real-world implementation of the proposed approach.

• Land and Housing Review
• /
• v.2 no.4
• /
• pp.407-413
• /
• 2011

This paper reports a novel structural health monitoring (SHM) technique for detecting de-bonding between a concrete beam and CFRP (Carbon Fiber Reinforced Polymer) sheet that is attached to the concrete surface. To achieve this, a multi-scale actuated sensing system with a self-sensing circuit using piezoelectric active sensors is applied to the CFRP laminated concrete beam structure. In this self-sensing based multi-scale actuated sensing, one scale provides a wide frequency-band structural response from the self-sensed impedance measurements and the other scale provides a specific frequency-induced structural wavelet response from the self-sensed guided wave measurement. To quantify the de-bonding levels, the supervised learning-based statistical pattern recognition was implemented by composing a two-dimensional (2D) plane using the damage indices extracted from the impedance and guided wave features.