• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.05a
• /
• pp.358-361
• /
• 2021

In order to improve the accuracy of small target detection more effectively, this paper proposes an improved single shot detector (SSD) target detection and recognition method based on cspdarknet53, which introduces lightweight ECA attention mechanism and Feature Pyramid Network (FPN). First, the original SSD backbone network is replaced with cspdarknet53 to enhance the learning ability of the network. Then, a lightweight ECA attention mechanism is added to the basic convolution block to optimize the network. Finally, FPN is used to gradually fuse the multi-scale feature maps used for detection in the SSD from the deep to the shallow layers of the network to improve the positioning accuracy and classification accuracy of the network. Experiments show that the proposed target detection algorithm has better detection accuracy, and it improves the detection accuracy especially for small targets.

• IEIE Transactions on Smart Processing and Computing
• /
• v.6 no.3
• /
• pp.183-192
• /
• 2017

k-nearest neighbor (K-NN) is a well-known classification algorithm, being feature space-based on nearest-neighbor training examples in machine learning. However, K-NN, as we know, is a lazy learning method. Therefore, if a K-NN-based system very much depends on a huge amount of history data to achieve an accurate prediction result for a particular task, it gradually faces a processing-time performance-degradation problem. We have noticed that many researchers usually contemplate only classification accuracy. But estimation speed also plays an essential role in real-time prediction systems. To compensate for this weakness, this paper proposes correlation coefficient-based clustering (CCC) aimed at upgrading the performance of K-NN by leveraging processing-time speed and plurality rule-based density (PRD) to improve estimation accuracy. For experiments, we used real datasets (on breast cancer, breast tissue, heart, and the iris) from the University of California, Irvine (UCI) machine learning repository. Moreover, real traffic data collected from Ojana Junction, Route 58, Okinawa, Japan, was also utilized to lay bare the efficiency of this method. By using these datasets, we proved better processing-time performance with the new approach by comparing it with classical K-NN. Besides, via experiments on real-world datasets, we compared the prediction accuracy of our approach with density peaks clustering based on K-NN and principal component analysis (DPC-KNN-PCA).

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2018.10a
• /
• pp.311-312
• /
• 2018

This paper proposes an image processing system based on the Convolution Neural Network technique. The image classification was performed using the composite neural network model and the images were classified with accuracy of 84% or more. The proposed system is implemented to operate on various platforms. When the system is used in the classification of images, the efficiency is higher because it is higher than the accuracy of the existing model.

• Proceedings of the Korea Institute of Convergence Signal Processing
• /
• 2000.12a
• /
• pp.145-148
• /
• 2000

Position-determining capacity is a very important condition in equipments for manufacturing semi-conductor or various instruments to measure physical displacement quantities of a moving object in submicron such as a distance of movement, direction, etc. and the accuracy of total system is influenced by detecting accuracy of these equipments. Therefore in this paper we have optically made up laser linear encoder based on optical diffraction principle to measure these displacement quantities and have processed optical signal using hardware-setup. In consequence we had acquired displacement for movement of scale using a diffraction grating by the accuracy of 0.5${\mu}{\textrm}{m}$ and had digitalized moving quantities of scale.

• Journal of Information Processing Systems
• /
• v.19 no.3
• /
• pp.370-376
• /
• 2023

Aiming at the problems of low image enhancement accuracy, long enhancement time and poor image quality in the traditional robot sequence motion image enhancement methods, an adaptive enhancement method for robot sequence motion image is proposed. The feature representation of the image was obtained by Karhunen-Loeve (K-L) transformation, and the nonlinear relationship between the robot joint angle and the image feature was established. The trajectory planning was carried out in the robot joint space to generate the robot sequence motion image, and an adaptive homomorphic filter was constructed to process the noise of the robot sequence motion image. According to the noise processing results, the brightness of robot sequence motion image was enhanced by using the multi-scale Retinex algorithm. The simulation results showed that the proposed method had higher accuracy and consumed shorter time for enhancement of robot sequence motion images. The simulation results showed that the image enhancement accuracy of the proposed method could reach 100%. The proposed method has important research significance and economic value in intelligent monitoring, automatic driving, and military fields.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.20 no.3
• /
• pp.243-253
• /
• 2002

These days, the application range of digital photogrammetry is being extended actively, so the great variety of research is being progressed about improvement of measurement accuracy. This method can acquisite a high analysis accuracy and can acquisite easily image. But, So this method require a much processing time and effort by increasing a data amount, as the automation of processing must be completed for efficiency improvement of processing. In this study, the automation of image processing stage is established for multi-image aquisition and analysis about a radio telescope. Also, the precise measurement and the accuracy on object were intended to be improved. As a result of this study, the accuracy on the object which demands precise measurement was enhanced from multi image analysis, and also the assembly accuracy of the object could be checked by precise measurement and curve fitting.

• Proceedings of the KIEE Conference
• /
• 2001.07e
• /
• pp.134-140
• /
• 2001

Parallel processing has the potential to be cost effectively used on computationally intense power system problems. But this technology is not still available is not only parallel computer but also parallel processing scheme. Testing these algorithms to ensure accuracy, and evaluation of their performance is also an issue. Although a significant amount of parallel algorithms of power system problem have been developed in last decade, actual testing on processor architectures lies in the beginning stages. This paper presents the parallel processing algorithm to supply the base being able to treat power flow by newton's method by the distributed memory type parallel computer. This method is to assign and to compute teared blocks of sparse matrix at each parallel processors. The testing to insure accuracy of developed method have been done on serial computer by trying to simulate a parallel environment.

• Journal of applied mathematics & informatics
• /
• v.27 no.1_2
• /
• pp.69-78
• /
• 2009

The accuracy of wavelet approximation at resolution h = $2^{-k}$ to a smooth function f is limited by O($h^M$), where M is the number of vanishing moments of the mother wavelet ${\psi}$; that is, the approximation order of wavelet approximation is M - 1. High accuracy points of wavelet approximation are of interest in some applications such as signal processing and numerical approximation. In this paper, we prove the scaling and translating properties of high accuracy points of wavelet approximation. To illustrate the results in this paper, we also present two examples of high accuracy points of wavelet approximation.

• 한국공간정보시스템학회:학술대회논문집
• /
• 2005.05a
• /
• pp.345-349
• /
• 2005

Massive amount of satellite image processing such asglobal/continental-level analysis and monitoring requires automated and speedy georegistration. There could be two major automated approaches: (1) rigid mathematical modeling using sensor model and ephemeris data; (2) heuristic co-registration approach with respect to existing reference image. In case of ETM+, the accuracy of the first approach is known as RMSE 250m, which is far below requested accuracy level for most of satellite image processing. On the other hands, the second approach is to find identical points between new image and reference image and use heuristic regression model for registration. The latter shows better accuracy but has problems with expensive computation. To improve efficiency of the coregistration approach, the author proposed a pre-qualified matching algorithm which is composed of feature extraction with canny operator and area matching algorithm with correlation coefficient. Throughout the pre-qualification approach, the computation time was significantly improved and make the registration accuracy is improved. A prototype was implemented and tested with the proposed algorithm. The performance test of 14 TM/ETM+ images in the U.S. showed: (1) average RMSE error of the approach was 0.47 dependent upon terrain and features; (2) the number average matching points were over 15,000; (3) the time complexity was 12 min per image with 3.2GHz Intel Pentium 4 and 1G Ram.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.66 no.9
• /
• pp.1402-1408
• /
• 2017

Recently, services for personal biometric data analysis based on real-time monitoring systems has been increasing and many of them have focused on recognition of emotions. In this paper, we propose a classification model to classify anxiety emotion using biometric data actually collected from people. We propose to deploy the support vector machine to build a classification model. In order to improve the classification accuracy, we propose two data pre-processing procedures, which are normalization and data deletion. The proposed algorithms are actually implemented based on Real-time Traffic Flow Measurement structure, which consists of data collection module, data preprocessing module, and creating classification model module. Our experiment results show that the proposed classification model can infers anxiety emotions of people with the accuracy of 65.18%. Moreover, the proposed model with the proposed pre-processing techniques shows the improved accuracy, which is 78.77%. Therefore, we can conclude that the proposed classification model based on the pre-processing process can improve the classification accuracy with lower computation complexity.