• Journal of the Korea Society of Computer and Information
• /
• v.10 no.4 s.36
• /
• pp.191-201
• /
• 2005

In this paper, we propose a method to calculate camera motion parameter, which is based on efficient invariant features irrelevant to the camera veiwpoint. As feature information in previous research is variant to camera viewpoint. information content is increased, therefore, extraction of accurate features is difficult. LM(Levenberg-Marquardt) method for camera extrinsic parameter converges on the goat value exactly, but it has also drawback to take long time because of minimization process by small step size. Therefore, in this paper, we propose the extracting method of invariant features to camera viewpoint and two-stage calculation method of camera motion parameter which enhances accuracy and convergent degree by using camera motion parameter by 2D homography to the initial value of LM method. The proposed method are composed of features extraction stage, matching stage and calculation stage of motion parameter. In the experiments, we compare and analyse the proposed method with existing methods by using various indoor images to demonstrate the superiority of the proposed algorithm.

• Journal of the Korea Society of Computer and Information
• /
• v.19 no.11
• /
• pp.63-72
• /
• 2014

In this paper, we propose a sign language recognition system using SVM and depth camera. Especially, we focus on the Korean sign language. For the sign language system, we suggest two methods, one in hand feature extraction stage and the other in recognition stage. Hand features are consisted of the number of fingers, finger length, radius of palm, and direction of the hand. To extract hand features, we use Distance Transform and make hand skeleton. This method is more accurate than a traditional method which uses contours. To recognize hand posture, we develop the decision tree with the hand features. For more accuracy, we use SVM to determine the threshold value in the decision tree. In the experimental results, we show that the suggested method is more accurate and faster when extracting hand features a recognizing hand postures.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.29 no.3
• /
• pp.531-539
• /
• 2019

This paper proposes the training features for malware family analysis and analyzes the multi-classification performance of ensemble models. We construct training data by extracting API and DLL information from malware executables and use Random Forest and XGBoost algorithms which are based on decision tree. API, API-DLL, and DLL-CM features for malware detection and family classification are proposed by analyzing frequently used API and DLL information from malware and converting high-dimensional features to low-dimensional features. The proposed feature selection method provides the advantages of data dimension reduction and fast learning. In performance comparison, the malware detection rate is 93.0% for Random Forest, the accuracy of malware family dataset is 92.0% for XGBoost, and the false positive rate of malware family dataset including benign is about 3.5% for Random Forest and XGBoost.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.2
• /
• pp.61-68
• /
• 2021

Recently, various technologies that use machine learning to classify malicious code have been studied. In order to enhance the effectiveness of machine learning, it is most important to extract properties to identify malicious codes and normal binaries. In this paper, we propose a feature extraction method for use in machine learning using recursive methods. The proposed method selects the final feature using recursive methods for individual features to maximize the performance of machine learning. In detail, we use the method of extracting the best performing features among individual feature at each stage, and then combining the extracted features. We extract features with the proposed method and apply them to machine learning algorithms such as Decision Tree, SVM, Random Forest, and KNN, to validate that machine learning performance improves as the steps continue.

• Smart Media Journal
• /
• v.10 no.1
• /
• pp.63-69
• /
• 2021

When a person sees a sentence and understands the sentence, the person understands the sentence by reminiscent of the main word in the sentence as an image. Text-to-image is what allows computers to do this associative process. The previous deep learning-based text-to-image model extracts text features using Convolutional Neural Network (CNN)-Long Short Term Memory (LSTM) and bi-directional LSTM, and generates an image by inputting it to the GAN. The previous text-to-image model uses basic embedding in text feature extraction, and it takes a long time to train because images are generated using several modules. Therefore, in this research, we propose a method of extracting features by using the attention mechanism, which has improved performance in the natural language processing field, for sentence embedding, and generating an image by inputting the extracted features into the GAN. As a result of the experiment, the inception score was higher than that of the model used in the previous study, and when judged with the naked eye, an image that expresses the features well in the input sentence was created. In addition, even when a long sentence is input, an image that expresses the sentence well was created.

• Journal of Internet Computing and Services
• /
• v.24 no.2
• /
• pp.11-18
• /
• 2023

Recently, in the field of Speech Emotion Recognition (SER), many studies have been conducted to improve accuracy using voice features and modeling. In addition to modeling studies to improve the accuracy of existing voice emotion recognition, various studies using voice features are being conducted. This paper, voice files are separated by time interval in a time series method, focusing on the fact that voice emotions are related to time flow. After voice file separation, we propose a model for classifying emotions of speech data by extracting speech features Mel, Chroma, zero-crossing rate (ZCR), root mean square (RMS), and mel-frequency cepstrum coefficients (MFCC) and applying them to a recurrent neural network model used for sequential data processing. As proposed method, voice features were extracted from all files using 'librosa' library and applied to neural network models. The experimental method compared and analyzed the performance of models of recurrent neural network (RNN), long short-term memory (LSTM) and gated recurrent unit (GRU) using the Interactive emotional dyadic motion capture Interactive Emotional Dyadic Motion Capture (IEMOCAP) english dataset.

• International conference on construction engineering and project management
• /
• 2022.06a
• /
• pp.877-885
• /
• 2022

Vision-based activity recognition has been widely attempted at construction sites to estimate productivity and enhance workers' health and safety. Previous studies have focused on extracting an individual worker's postural information from sequential image frames for activity recognition. However, various trades of workers perform different tasks with similar postural patterns, which degrades the performance of activity recognition based on postural information. To this end, this research exploited a concept of human-object interaction, the interaction between a worker and their surrounding objects, considering the fact that trade workers interact with a specific object (e.g., working tools or construction materials) relevant to their trades. This research developed an approach to understand the context from sequential image frames based on four features: posture, object, spatial features, and temporal feature. Both posture and object features were used to analyze the interaction between the worker and the target object, and the other two features were used to detect movements from the entire region of image frames in both temporal and spatial domains. The developed approach used convolutional neural networks (CNN) for feature extractors and activity classifiers and long short-term memory (LSTM) was also used as an activity classifier. The developed approach provided an average accuracy of 85.96% for classifying 12 target construction tasks performed by two trades of workers, which was higher than two benchmark models. This experimental result indicated that integrating a concept of the human-object interaction offers great benefits in activity recognition when various trade workers coexist in a scene.

• Current Optics and Photonics
• /
• v.8 no.3
• /
• pp.215-224
• /
• 2024

In medium- and long-range optical imaging systems, atmospheric turbulence causes blurring and distortion of images, resulting in loss of image information. An image-restoration method based on an adaptive feature-fusion multi-input-multi-output (MIMO) dense U-shaped network (Unet) is proposed, to restore a single image degraded by atmospheric turbulence. The network's model is based on the MIMO-Unet framework and incorporates patch-embedding shallow-convolution modules. These modules help in extracting shallow features of images and facilitate the processing of the multi-input dense encoding modules that follow. The combination of these modules improves the model's ability to analyze and extract features effectively. An asymmetric feature-fusion module is utilized to combine encoded features at varying scales, facilitating the feature reconstruction of the subsequent multi-output decoding modules for restoration of turbulence-degraded images. Based on experimental results, the adaptive feature-fusion MIMO dense U-shaped network outperforms traditional restoration methods, CMFNet network models, and standard MIMO-Unet network models, in terms of image-quality restoration. It effectively minimizes geometric deformation and blurring of images.

• Journal of the Institute of Convergence Signal Processing
• /
• v.24 no.4
• /
• pp.221-228
• /
• 2023

Physical and mental changes caused by emotions can affect various behaviors, such as driving or learning behavior. Therefore, recognizing these emotions is a very important task because it can be used in various industries, such as recognizing and controlling dangerous emotions while driving. In this paper, we attempted to solve the emotion recognition task by implementing a multimodal model that recognizes emotions using both audio and video data from different domains. After extracting voice from video data using RAVDESS data, features of voice data are extracted through a model using 2D-CNN. In addition, the video data features are extracted using a slowfast feature extractor. And the information contained in the audio and video data, which have different domains, are combined into one feature that contains all the information. Afterwards, emotion recognition is performed using the combined features. Lastly, we evaluate the conventional methods that how to combine results from models and how to vote two model's results and a method of unifying the domain through feature extraction, then combining the features and performing classification using a classifier.

• Journal of IKEEE
• /
• v.28 no.3
• /
• pp.352-359
• /
• 2024

This paper proposes a CNN-ViT hybrid model that automatically evaluates the aesthetic quality of images by combining local and global features. In this approach, CNN is used to extract local features such as color and object placement, while ViT is employed to analyze the aesthetic value of the image by reflecting global features. Color composition is derived by extracting the primary colors from the input image, creating a color palette, and then passing it through the CNN. The Rule of Thirds is quantified by calculating how closely objects in the image are positioned near the thirds intersection points. These values provide the model with critical information about the color balance and spatial harmony of the image. The model then analyzes the relationship between these factors to predict scores that align closely with human judgment. Experimental results on the AADB image database show that the proposed model achieved a Spearman's Rank Correlation Coefficient (SRCC) of 0.716, indicating more consistent rank predictions, and a Pearson Correlation Coefficient (LCC) of 0.72, which is 2~4% higher than existing models.