• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.4
• /
• pp.1807-1823
• /
• 2016

Local Binary Pattern (LBP) and its variants have powerful discriminative capabilities but most of them just consider each LBP code independently. In this paper, we propose sub oriented histograms of LBP for smoke detection and image classification. We first extract LBP codes from an image, compute the gradient of LBP codes, and then calculate sub oriented histograms to capture spatial relations of LBP codes. Since an LBP code is just a label without any numerical meaning, we use Hamming distance to estimate the gradient of LBP codes instead of Euclidean distance. We propose to use two coordinates systems to compute two orientations, which are quantized into discrete bins. For each pair of the two discrete orientations, we generate a sub LBP code map from the original LBP code map, and compute sub oriented histograms for all sub LBP code maps. Finally, all the sub oriented histograms are concatenated together to form a robust feature vector, which is input into SVM for training and classifying. Experiments show that our approach not only has better performance than existing methods in smoke detection, but also has good performance in texture classification.

• Journal of the Institute of Convergence Signal Processing
• /
• v.8 no.4
• /
• pp.285-292
• /
• 2007

SoC which deploys software modules as well as hardware IPs on a single chip is a major revolution taking place in the implementation of a system design, and high-level synthesis is an important process of SoC design methodology. Recently, SPARK parallelizing high-level synthesis software tool has been developed. It takes a behavioral ANSI-C code as an input, schedules it using code motion and various code transformations, and then finally generates synthesizable RTL VHDL code. Although SPARK employs various loop transformation algorithms, the synthesis results generated by SPARK are not acceptable for basic signal and image processing algorithms with nested loop. In this paper we propose a SoC platform with an application-specific PLD targeting local operations which are feature of many loop algorithms used in signal and image processing, and demonstrate design process which maps behavioral specification with nested loops written in a high-level language (ANSI-C) onto 2D systolic array. Finally the derived systolic array is implemented on the proposed application-specific PLD of SoC platform.

• Journal of Korean Society for Geospatial Information Science
• /
• v.17 no.3
• /
• pp.13-21
• /
• 2009

Digital map should have the up-to-dateness as well as the accuracy to perform a role as the national spatial data. As digital mapping process require aerial photograph, surveying, and field working, it consumes a lot of time and cost. So there is a limit to maintain the up-to-dateness. If we updates the digital map frequently by using the as-built drawings, we can prevent the waste of national budget by reuse of existing drawings and make accuracy updates from existing survey results. In spite of this advantages, due to insufficiency of CAD drawing standard, inconsistency of file types of as-built drawings and digital maps, and topology relations between input features and original features, so the frequent updates using the as-built drawings is on the difficult situation to perform. In this research, first, CAD features extracted from as-built drawings land the new/update whether original features exist or not and generate topology from spatial relation of features. Second, suggest the efficient partial-update-plan performing integrity test. As a result, guarantee the accuracy and the up-to-dateness of digital map.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.6 no.3
• /
• pp.73-82
• /
• 2003

This paper proposes a work procedure for generalizing large-scale digital maps ver. 2.0(1/5,000) into a small-scale digital map(1/25,000). Unlike a existent digital map, the digital map ver. 2.0 has a variety of attribute data as well as graphic data. To perform an efficient map generalization with these structural properties, we establish a work procedure as follow; firstly, delete layers which don't exist in small-scale digital map's feature code, and secondly, generalize features which have been classified into 8 layers, and finally merge 8 layers which have been generalized into 1 layer. Therefore, we expect that a work procedure which is proposed in this paper will play a fundamental role in automated generalization system and will contribute to small-scale digital mapping and thematic mapping.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.4
• /
• pp.246-254
• /
• 2020

In this paper, we propose a learning based compressive sensing algorithm for the purpose of side scan sonar image denoising. The proposed method is based on Iterative Shrinkage and Thresholding Algorithm (ISTA) framework and incorporates a powerful strategy that reinforces the non-linearity of deep learning network for improved performance. The proposed method consists of three essential modules. The first module consists of a non-linear transform for input and initialization while the second module contains the ISTA block that maps the input features to sparse space and performs inverse transform. The third module is to transform from non-linear feature space to pixel space. Superiority in noise removal and memory efficiency of the proposed method is verified through various experiments.

• Journal of KIISE:Software and Applications
• /
• v.31 no.2
• /
• pp.189-202
• /
• 2004

In this paper, we propose a new salient region detection method in an image. The algorithm is based on the characteristics of human's bottom-up visual attention. Several features known to influence human visual attention like color, intensity and etc. are extracted from the each regions of an image. These features are then converted to importance values for each region using its local competition function and are combined to produce a saliency map, which represents the saliency at every location in the image by a scalar quantity, and guides the selection of attended locations, based on the spatial distribution of saliency region of the image in relation to its Perceptual importance. Results shown indicate that the calculated Saliency Maps correlate well with human perception of visually important regions.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.18 no.3
• /
• pp.106-118
• /
• 2019

In recent years, research and development on image object recognition using artificial intelligence have been actively carried out, and it is expected to be used for road maintenance. Among them, artificial intelligence models for object detection of road surface are continuously introduced. In order to develop such object recognition algorithms, a backbone network that extracts feature maps is essential. In this paper, we will discuss how to select the appropriate neural network. To accomplish it, we compared with 4 different deep neural networks using 6,000 road surface damage images. Based on three evaluation methods for analyzing characteristics of neural networks, we propose a method to determine optimal neural networks. In addition, we improved the performance through optimal tuning of hyper-parameters, and finally developed a light backbone network that can achieve 85.9% accuracy of road surface damage classification.

• Steel and Composite Structures
• /
• v.44 no.2
• /
• pp.241-254
• /
• 2022

The cables in a cable-stayed bridge are critical load-carrying parts. The potential damage to cables should be identified early to prevent disasters. In this study, an efficient deep learning model is proposed for the damage identification of cables using both a multi-layer perceptron (MLP) and a graph neural network (GNN). Datasets are first generated using the practical advanced analysis program (PAAP), which is a robust program for modeling and analyzing bridge structures with low computational costs. The model based on the MLP and GNN can capture complex nonlinear correlations between the vibration characteristics in the input data and the cable system damage in the output data. Multiple hidden layers with an activation function are used in the MLP to expand the original input vector of the limited measurement data to obtain a complete output data vector that preserves sufficient information for constructing the graph in the GNN. Using the gated recurrent unit and set2set model, the GNN maps the formed graph feature to the output cable damage through several updating times and provides the damage results to both the classification and regression outputs. The model is fine-tuned with the original input data using Adam optimization for the final objective function. A case study of an actual cable-stayed bridge was considered to evaluate the model performance. The results demonstrate that the proposed model provides high accuracy (over 90%) in classification and satisfactory correlation coefficients (over 0.98) in regression and is a robust approach to obtain effective identification results with a limited quantity of input data.

• 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%.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.7 no.10
• /
• pp.935-943
• /
• 2017

CNN (Convolution neural network), which is used for image classification and speech recognition among neural networks learning based on positive data, has been continuously developed to have a high performance structure to date. There are many difficulties to utilize in an embedded system with limited resources. Therefore, we use GPU (General-Purpose Computing on Graphics Processing Units), which is used for general-purpose operation of GPU to solve the problem because we use pre-learned weights but there are still limitations. Since CNN performs simple and iterative operations, the computation speed varies greatly depending on the thread allocation and utilization method in the Single Instruction Multiple Thread (SIMT) based GPGPU. To solve this problem, there is a thread that needs to be relaxed when performing Convolution and Pooling operations with threads. The remaining threads have increased the operation speed by using the method used in the following feature maps and kernel calculations.