• Smart Media Journal
• /
• v.12 no.1
• /
• pp.9-16
• /
• 2023

Recently, a convolutional neural network (CNN) based system is being developed to overcome the limitations of human resources in the apple quality classification of farmhouse. However, since convolutional neural networks receive only images of the same size, preprocessing such as sampling may be required, and in the case of oversampling, information loss of the original image such as image quality degradation and blurring occurs. In this paper, in order to minimize the above problem, to generate a image patch based graph of an original image and propose a random walk-based positional encoding method to apply the graph transformer model. The above method continuously learns the position embedding information of patches which don't have a positional information based on the random walk algorithm, and finds the optimal graph structure by aggregating useful node information through the self-attention technique of graph transformer model. Therefore, it is robust and shows good performance even in a new graph structure of random node order and an arbitrary graph structure according to the location of an object in an image. As a result, when experimented with 5 apple quality datasets, the learning accuracy was higher than other GNN models by a minimum of 1.3% to a maximum of 4.7%, and the number of parameters was 3.59M, which was about 15% less than the 23.52M of the ResNet18 model. Therefore, it shows fast reasoning speed according to the reduction of the amount of computation and proves the effect.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.5
• /
• pp.2607-2627
• /
• 2017

Nonnegative matrix factorization (NMF) has received considerable attention due to its effectiveness of reducing high dimensional data and importance of producing a parts-based image representation. Most of existing NMF variants attempt to address the assertion that the observed data distribute on a nonlinear low-dimensional manifold. However, recent research results showed that not only the observed data but also the features lie on the low-dimensional manifolds. In addition, a few hard priori label information is available and thus helps to uncover the intrinsic geometrical and discriminative structures of the data space. Motivated by the two aspects above mentioned, we propose a novel algorithm to enhance the effectiveness of image representation, called Dual graph-regularized Constrained Nonnegative Matrix Factorization (DCNMF). The underlying philosophy of the proposed method is that it not only considers the geometric structures of the data manifold and the feature manifold simultaneously, but also mines valuable information from a few known labeled examples. These schemes will improve the performance of image representation and thus enhance the effectiveness of image classification. Extensive experiments on common benchmarks demonstrated that DCNMF has its superiority in image classification compared with state-of-the-art methods.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.36 no.3
• /
• pp.135-152
• /
• 2018

The purpose of this study is to propose an optimal fusion method of aerial multi - sensor data to improve the accuracy of land cover classification. Recently, in the fields of environmental impact assessment and land monitoring, high-resolution image data has been acquired for many regions for quantitative land management using aerial multi-sensor, but most of them are used only for the purpose of the project. Hyperspectral sensor data, which is mainly used for land cover classification, has the advantage of high classification accuracy, but it is difficult to classify the accurate land cover state because only the visible and near infrared wavelengths are acquired and of low spatial resolution. Therefore, there is a need for research that can improve the accuracy of land cover classification by fusing hyperspectral sensor data with multispectral sensor and aerial laser sensor data. As a fusion method of aerial multisensor, we proposed a pixel ratio adjustment method, a band accumulation method, and a spectral graph adjustment method. Fusion parameters such as fusion rate, band accumulation, spectral graph expansion ratio were selected according to the fusion method, and the fusion data generation and degree of land cover classification accuracy were calculated by applying incremental changes to the fusion variables. Optimal fusion variables for hyperspectral data, multispectral data and aerial laser data were derived by considering the correlation between land cover classification accuracy and fusion variables.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.2
• /
• pp.148-155
• /
• 2014

In this paper a commute time embedding is implemented by organizing patches according to the graph-based metric, and its properties are investigated via changing the number of nodes on the graph.. It is shown that manifold embedding methods generate the intrinsic geometric structures when waveforms such as speech or music instrumental sound signals are embedded on the low dimensional Euclidean space. Basically manifold embedding algorithms only project the training samples on the graph into an embedding subspace but can not generalize the learning results to test samples. They are very effective for data clustering but are not appropriate for classification or recognition. In this paper a commute time guided transform is adopted to enhance the generalization ability and its performance is analyzed by applying it to the classification of 6 kinds of music instrumental sounds.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.18 no.6
• /
• pp.1269-1276
• /
• 2023

Crop estimation is essential for the multinational meal and powerful demand due to its numerous aspects like soil, rain, climate, atmosphere, and their relations. The consequence of climate shift impacts the farming yield products. We operate the dataset with temperature, rainfall, humidity, etc. The current research focuses on feature selection with multifarious classifiers to assist farmers and agriculturalists. The crop yield estimation utilizing the feature selection approach is 96% accuracy. Feature selection affects a machine learning model's performance. Additionally, the performance of the current graph classifier accepts 81.5%. Eventually, the random forest regressor without feature selections owns 78% accuracy and the decision tree regressor without feature selections retains 67% accuracy. Our research merit is to reveal the experimental results of with and without feature selection significance for the proposed ten algorithms. These findings support learners and students in choosing the appropriate models for crop classification studies.

• Korean Journal of Computational Design and Engineering
• /
• v.6 no.2
• /
• pp.78-88
• /
• 2001

In order to adopt feature-based parametric modeling, CAD/CAM applications must have a geometric constraint solver that can handle a large set of geometric configurations efficiently and robustly. In this paper, we describe a graph constructive approach to solving geometric constraint problems. Usually, a graph constructive approach is efficient, however it has its limitation in scope; it cannot handle ruler-and-compass non-constructible configurations and under-constrained problems. To overcome these limitations. we propose an algorithm that isolates ruler-and-compass non-constructible configurations from ruler-and-compass constructible configurations and applies numerical calculation methods to solve them separately. This separation can maximize the efficiency and robustness of a geometric constraint solver. Moreover, the solver can handle under-constrained problems by classifying under-constrained subgraphs to simplified cases by applying classification rules. Then, it decides the calculating sequence of geometric entities in each classified case and calculates geometric entities by adding appropriate assumptions or constraints. By extending the clustering types and defining several rules, the proposed approach can overcome limitations of previous graph constructive approaches which makes it possible to develop an efficient and robust geometric constraint solver.

• The Journal of Society for e-Business Studies
• /
• v.26 no.1
• /
• pp.29-41
• /
• 2021

Recently, there are increasing attempts to utilize deep learning methodology in the fashion industry. Accordingly, research dealing with various fashion-related problems have been proposed, and superior performances have been achieved. However, the studies for fashion style classification have not reflected the characteristics of the fashion style that one outfit can include multiple styles simultaneously. Therefore, we aim to solve the multi-label classification problem by utilizing the dependencies between the styles. A multi-label recognition model based on a graph convolution network is applied to detect and explore fashion styles' dependencies. Furthermore, we accelerate model training and improve the model's performance through transfer learning. The proposed model was verified by a dataset collected from social network services and outperformed baselines.

• Journal of the Economic Geographical Society of Korea
• /
• v.15 no.3
• /
• pp.343-357
• /
• 2012

The Seoul metropolitan subway system can be represented by a graph which consists of nodes and edges. In this paper, we study classification of subway stations and trip behaviour of subway passengers through partitioning the graph of the subway system into roughly equal groups. A weight of each edge of the graph is set to the number of passengers who pass the edge, where the number of passengers is extracted from the transportation card transaction database. Since the graph partitioning problem is NP-complete, we propose a heuristic algorithm to partition the subway graph. The heuristic algorithm uses one of two alternative objective functions, one of which is to minimize the sum of weights of edges connecting nodes in different groups and the other is to maximize the ratio of passengers who get on the subway train at one subway station and get off at another subway station in the same group to the total subway passengers. In the experimental results, we illustrate the subway stations and edges in each group by color on a map and analyze the trip behaviour of subway passengers by the group origin-destination matrix.

• International journal of advanced smart convergence
• /
• v.12 no.2
• /
• pp.34-46
• /
• 2023

A growing number of large scale knowledge graphs raises several issues how knowledge graph data can be organized, discovered, and integrated efficiently. We present a novel semantic-based mashup platform for contents convergence which consists of acquisition, RDF storage, ontology learning, and mashup subsystems. This platform servers a basis for developing other more sophisticated applications required in the area of knowledge big data. Moreover, this paper proposes an entity matching method using graph convolutional network techniques as a preliminary work for automatic classification and discovery on knowledge big data. Using real DBP15K and SRPRS datasets, the performance of our method is compared with some existing entity matching methods. The experimental results show that the proposed method outperforms existing methods due to its ability to increase accuracy and reduce training time.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.18 no.3
• /
• pp.565-572
• /
• 1994

A method for the classification of kinematic chains according to the similarity in their structures is proposed. Classifcation code is defined from the contracted graph of the kinematic chain. This method of classifying kinematic chains can be effectively used for the systematic enumeration of structurally distinct kinematic chains given the number of links and degrees of freedom of the kinematic chains. Two separate steps for the enumeration are developed in the study. The first step is to generated all the possible classification codes and the next step is to generate individual kinematic chains belonging to each classification code generated. Using this two step procedure, kinematic chains up to 12 links are successfully enumerated in the present study. It is concluded that the two step method can be efficiently used for the type synthesis of mechanisms.