• Journal of the Korea Society of Computer and Information
• /
• v.26 no.1
• /
• pp.27-35
• /
• 2021

In this paper, we propose a new stacking ensemble framework for deep learning models which reflects the distribution of label embeddings. Our ensemble framework consists of two phases: training the baseline deep learning classifier, and training the sub-classifiers based on the clustering results of label embeddings. Our framework aims to divide a multi-class classification problem into small sub-problems based on the clustering results. The clustering is conducted on the label embeddings obtained from the weight of the last layer of the baseline classifier. After clustering, sub-classifiers are constructed to classify the sub-classes in each cluster. From the experimental results, we found that the label embeddings well reflect the relationships between classification labels, and our ensemble framework can improve the classification performance on a CIFAR 100 dataset.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.22 no.6
• /
• pp.1-16
• /
• 2023

Accurate traffic information prediction is considered to be one of the most important aspects of intelligent transport systems(ITS), as it can be used to guide users of transportation facilities to avoid congested routes. Various deep learning models have been developed for accurate traffic prediction. Recently, ensemble techniques have been utilized to combine the strengths and weaknesses of various models in various ways to improve prediction accuracy and stability. Therefore, in this study, we developed and evaluated a traffic information prediction model using various deep learning models, and evaluated the performance of the developed deep learning models as a stacking ensemble. The individual models showed error rates within 10% for traffic volume prediction and 3% for speed prediction. The ensemble model showed higher accuracy compared to other models when no cross-validation was performed, and when cross-validation was performed, it showed a uniform error rate in long-term forecasting.

• Journal of Korea Multimedia Society
• /
• v.25 no.8
• /
• pp.1212-1223
• /
• 2022

Skin lesions are common diseases that range from skin rashes to skin cancer, which can lead to death. Note that early diagnosis of skin diseases can be important because early diagnosis of skin diseases considerably can reduce the course of treatment and the harmful effect of the disease. Recently, the development of computer-aided diagnosis (CAD) systems based on artificial intelligence has been actively made for the early diagnosis of skin diseases. In a typical CAD system, the accurate classification of skin lesion types is of great importance for improving the diagnosis performance. Motivated by this, we propose a novel deep ensemble classification with multi-scale attention networks. The proposed deep ensemble networks are jointly trained using a single loss function in an end-to-end manner. In addition, the proposed deep ensemble network is equipped with a multi-scale attention mechanism and segmentation information of the original skin input image, which improves the classification performance. To demonstrate our method, the publicly available human skin disease dataset (HAM 10000) and the private animal skin lesion dataset were used for the evaluation. Experiment results showed that the proposed methods can achieve 97.8% and 81% accuracy on each HAM10000 and animal skin lesion dataset. This research work would be useful for developing a more reliable CAD system which helps doctors early diagnose skin diseases.

• Journal of Information Processing Systems
• /
• v.19 no.5
• /
• pp.673-687
• /
• 2023

Convolutional neural network-based deep learning technology is the most commonly used in image identification, but it requires large-scale data for training. Therefore, application in specific fields in which data acquisition is limited, such as in the military, may be challenging. In particular, the identification of ground weapon systems is a very important mission, and high identification accuracy is required. Accordingly, various studies have been conducted to achieve high performance using small-scale data. Among them, the ensemble method, which achieves excellent performance through the prediction average of the pre-trained models, is the most representative method; however, it requires considerable time and effort to find the optimal combination of ensemble models. In addition, there is a performance limitation in the prediction results obtained by using an ensemble method. Furthermore, it is difficult to obtain the ensemble effect using models with imbalanced classification accuracies. In this paper, we propose a transfer learning-based feature fusion technique for heterogeneous models that extracts and fuses features of pre-trained heterogeneous models and finally, fine-tunes hyperparameters of the fully connected layer to improve the classification accuracy. The experimental results of this study indicate that it is possible to overcome the limitations of the existing ensemble methods by improving the classification accuracy through feature fusion between heterogeneous models based on transfer learning.

• ETRI Journal
• /
• v.41 no.6
• /
• pp.750-759
• /
• 2019

Presently, the extraction of hand-crafted features is still the dominant method in radar emitter recognition. To solve the complicated problems of selection and updation of empirical features, we present a novel automatic feature extraction structure based on deep learning. In particular, a convolutional neural network (CNN) is adopted to extract high-level abstract representations from the time-frequency images of emitter signals. Thus, the redundant process of designing discriminative features can be avoided. Furthermore, to address the performance degradation of a single platform, we propose the construction of an ensemble learning-based architecture for multi-platform fusion recognition. Experimental results indicate that the proposed algorithms are feasible and effective, and they outperform other typical feature extraction and fusion recognition methods in terms of accuracy. Moreover, the proposed structure could be extended to other prevalent ensemble learning alternatives.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.11
• /
• pp.29-38
• /
• 2022

This paper proposes a system for classifying gait types using an ensemble deep learning network for gait data measured by a smart insole equipped with multi-sensors. The gait type classification system consists of a part for normalizing the data measured by the insole, a part for extracting gait features using a deep learning network, and a part for classifying the gait type by inputting the extracted features. Two kinds of gait feature maps were extracted by independently learning networks based on CNNs and LSTMs with different characteristics. The final ensemble network classification results were obtained by combining the classification results. For the seven types of gait for adults in their 20s and 30s: walking, running, fast walking, going up and down stairs, and going up and down hills, multi-sensor data was classified into a proposed ensemble network. As a result, it was confirmed that the classification rate was higher than 90%.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.7
• /
• pp.37-44
• /
• 2021

Automatic classification of brain MRI images play an important role in early diagnosis of brain tumors. In this work, we present a deep learning-based brain tumor classification model in MRI images using ensemble of deep features. In our proposed framework, three different deep features from brain MR image are extracted using three different pre-trained models. After that, the extracted deep features are fed to the classification module. In the classification module, the three different deep features are first fed into the fully-connected layers individually to reduce the dimension of the features. After that, the output features from the fully-connected layers are concatenated and fed into the fully-connected layer to predict the final output. To evaluate our proposed model, we use openly accessible brain MRI dataset from web. Experimental results show that our proposed model outperforms other machine learning-based models.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.20 no.4
• /
• pp.101-106
• /
• 2020

In the current medical field, ultrasound diagnosis can be said to be the same as a stethoscope in the past. However, due to the nature of ultrasound, it has the disadvantage that the prediction of results is uncertain depending on the skill level of the examiner. Therefore, this paper aims to improve the accuracy of liver lesion detection during ultrasound examination based on deep learning technology to solve this problem. In the proposed paper, we compared the accuracy of lesion classification using a CNN model and an ensemble model. As a result of the experiment, it was confirmed that the classification accuracy in the CNN model averaged 82.33% and the ensemble model averaged 89.9%, about 7% higher. Also, it was confirmed that the ensemble model was 0.97 in the average ROC curve, which is about 0.4 higher than the CNN model.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.23 no.2
• /
• pp.139-146
• /
• 2020

The camouflage pattern was difficult to distinguish from the surrounding background, so it was difficult to classify the object and the background image when the color image is used as the training data of deep-learning. In this paper, we proposed a red-green color blindness image transformation method using the principle that people of red-green blindness distinguish green color better than ordinary people. Experimental results show that the camouflage soldier's recognition performance improved by proposed a deep learning model of the ensemble technique using the imitated red-green-blind image data and the original color image data.

• The Journal of the Acoustical Society of Korea
• /
• v.42 no.4
• /
• pp.305-312
• /
• 2023

Physics-Informed Neural Network (PINN) is used to invert bubble size distributions from attenuation losses. By considering a linear system for the bubble population inversion, Adaptive Learned Iterative Shrinkage Thresholding Algorithm (Ada-LISTA), which has been solved linear systems in image processing, is used as a neural network architecture in PINN. Furthermore, a regularization based on the linear system is added to a loss function of PINN and it makes a PINN have better generalization by a solution satisfying the bubble physics. To evaluate an uncertainty of bubble estimation, deep ensemble is adopted. 20 Ada-LISTAs with different initial values are trained using the same training dataset. During test with attenuation losses different from those in the training dataset, the bubble size distribution and corresponding uncertainty are indicated by average and variance of 20 estimations, respectively. Deep ensemble Ada-LISTA demonstrate superior performance in inverting bubble size distributions than the conventional convex optimization solver of CVX.