• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.3
• /
• pp.1477-1491
• /
• 2017

Inspired by the multi-scale nature of hippocampal place cells, a biologically inspired model based on a multi-scale spatial representation for goal-directed navigation is proposed in order to achieve robotic spatial cognition and autonomous navigation. First, a map of the place cells is constructed in different scales, which is used for encoding the spatial environment. Then, the firing rate of the place cells in each layer is calculated by the Gaussian function as the input of the Q-learning process. The robot decides on its next direction for movement through several candidate actions according to the rules of action selection. After several training trials, the robot can accumulate experiential knowledge and thus learn an appropriate navigation policy to find its goal. The results in simulation show that, in contrast to the other two methods(G-Q, S-Q), the multi-scale model presented in this paper is not only in line with the multi-scale nature of place cells, but also has a faster learning potential to find the optimized path to the goal. Additionally, this method also has a good ability to complete the goal-directed navigation task in large space and in the environments with obstacles.

• The Journal of the Korea Contents Association
• /
• v.7 no.6
• /
• pp.52-59
• /
• 2007

This paper presents a greedy method for user-steered mesh segmentation, which is based on the merging priority metric defined for representing the geometric properties of meaningful parts. The priority metric is a weighted function composed of five geometric parameters: distribution of Gaussian map, boundary path concavity, boundary path length, cardinality, and segmentation resolution. This scheme can be extended without any modification only by defining more geometric parameters and adding them. Our experimental results show that the shapes of segmented parts can be controlled by setting up the weight values of geometric parameters.

• Proceedings of the IEEK Conference
• /
• 2000.07a
• /
• pp.478-481
• /
• 2000

In this paper, we propose a joint coding system for still images using source coding and powerful error correcting code schemes. Our system comprises an LVQ (lattice vector quantization) source coding for wavelet transformed images and turbo coding for channel coding. The parameters of the image encoder and channel encoder have been optimized for an n-D (dimension) cubic lattice (D$_{n}$, Z$_{n}$), parallel concatenation fur two simple RSC (recursive systematic convolutional code) and an interleaver. For decoding the received image in the case of the AWGN (additive white gaussian noise) channel, we used an iterative joint source-channel decoding algorithm for a SISO (soft-input soft-output) MAP (maximum a posteriori) module. The performance of transmission system has been evaluated in the PSNR, BER and iteration times. A very small degradation of the PSNR and an improvement in BER were compared to a system without joint source-channel decoding at the input of the receiver.ver.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.21 no.7
• /
• pp.1696-1705
• /
• 1996

The theory of stack filtering, which is a generalization of median filtering, is used to the detection of intensity edges in noisey images. The proposed approach, called the Difference of Estimates(DoE) approach, is a new formulation of a morphological scheme which has been very sensitive to impulse noise. In this approach, stack filters are applied to a noisy image to obtain local estimates of the dilated and eroded versions of the noise-free image. Thresholding the difference between these two estimates yields the binary edge map. We find that this approach yields results comparable to those obtained with the Canny operator for images with additive Gaussian noise, burt works much better when the noise is impulsive.

• Transactions on Electrical and Electronic Materials
• /
• v.13 no.1
• /
• pp.10-15
• /
• 2012

In conventional, skin detection methods using for skin color definitions is based on prior knowledge. By experimentation, the threshold value for dividing the background from the skin region is determined subjectively. A drawback of such techniques is that their performance is dependent on a threshold value which is estimated from repeated experiments. To overcome this, the present paper introduces a skin region detection method. This method uses a histogram approximation based on the mean shift algorithm. This proposed method applies the mean shift procedure to a histogram of a skin map of the input image. It is generated by comparing with the standard skin colors in the $C_bC_r$ color space. It divides the background from the skin region by selecting the maximum value according to the brightness level. As the histogram has the form of a discontinuous function. It is accumulated according to the brightness values of the pixels. It is then, approximated by a Gaussian mixture model (GMM) using the Bezier curve technique. Thus, the proposed method detects the skin region using the mean shift procedure to determine a maximum value. Rather than using a manually selected threshold value, as in existing techniques this becomes the dividing point. Experiments confirm that the new procedure effectively detects the skin region.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2015.10a
• /
• pp.1579-1580
• /
• 2015

본 논문에서는 3차원 메쉬 모델의 중요 영역을 표현하는 메쉬 돌출맵(mesh saliency map)을 생성하기 위하여 다중 스케일 평균 곡률 (multi-scale mean curvature)을 기반으로 정의된 전역 희소치(global rarity)를 이용하는 방법을 제안한다. 제안 방법에서는 우선, 메쉬 모델의 지역 영역 특성을 정의하기 위하여 기존 관련 연구들에서 많이 사용하고 있는 가우시안 가중치 평균곡률(Gaussian-weighted mean curvature)을 5단계 서로 다른 스케일에서 정의하고, 메쉬의 각 정점(vertex)에 대하여 중심주변 연산자(center-surround operator)를 적용하여 5단계 지역 돌출특성(local saliency)을 정의한다. 주어진 메쉬 모델의 전역 희소치를 구하기 위하여 메쉬의 모든 정점쌍 (vertex pair)에 대하여 5단계 지역 돌출 특성 공간에서의 거리를 계산하고, 각 정점별로 5단계 지역 돌출 특성 공간에서의 다른 정점과의 거리의 합으로 전역 희소치를 정의한다. 이러한 전역 희소치를 각 정점의 메쉬 돌출치로 정의한다. 서로 다른 형태의 3차원 모델에 대하여 제안방법에 의한 메쉬 돌출맵과 지역 특성만을 고려한 기존 메쉬 돌출맵을 생성하여 중요 영역 표현 결과를 비교 분석한다.

• ETRI Journal
• /
• v.44 no.5
• /
• pp.805-815
• /
• 2022

The model-based evolutionary algorithms are divided into three groups: estimation of distribution algorithms, inverse modeling, and surrogate modeling. Existing inverse modeling is mainly applied to solve multi-objective optimization problems and is not suitable for many-objective optimization problems. Some inversed-model techniques, such as the inversed-model of multi-objective evolutionary algorithm, constructed from the Pareto front (PF) to the Pareto solution on nondominated solutions using a random grouping method and Gaussian process, were introduced. However, some of the most efficient inverse models might be eliminated during this procedure. Also, there are challenges, such as the presence of many local PFs and developing poor solutions when the population has no evident regularity. This paper proposes inverse modeling using random forest regression and uniform reference points that map all nondominated solutions from the objective space to the decision space to solve many-objective optimization problems. The proposed algorithm is evaluated using the benchmark test suite for evolutionary algorithms. The results show an improvement in diversity and convergence performance (quality indicators).

• Journal of information and communication convergence engineering
• /
• v.21 no.3
• /
• pp.208-215
• /
• 2023

Deep learning techniques provide powerful solutions to several pattern-recognition problems, including Raman spectral classification. However, these networks require large amounts of labeled data to perform well. Labeled data, which are typically obtained in a laboratory, can potentially be alleviated by data augmentation. This study investigated various data augmentation techniques and applied multiple deep learning methods to Raman spectral classification. Raman spectra yield fingerprint-like information about chemical compositions, but are prone to noise when the particles of the material are small. Five augmentation models were investigated to build robust deep learning classifiers: weighted sums of spectral signals, imitated chemical backgrounds, extended multiplicative signal augmentation, and generated Gaussian and Poisson-distributed noise. We compared the performance of nine state-of-the-art convolutional neural networks with all the augmentation techniques. The LeNet5 models with background noise augmentation yielded the highest accuracy when tested on real-world Raman spectral classification at 88.33% accuracy. A class activation map of the model was generated to provide a qualitative observation of the results.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.277-277
• /
• 2021

합류부는 인공수로 또는 자연하천에서 흔히 존재하며 매우 복잡한 흐름 구조가 발생하는 곳이다. 특히 본류와 지류의 유속장의 차이에 따라 발생하는 전단층은 흐름과 물질이 혼합되는 경계면이 되며, 흐름 구조가 전단층을 따라 발달한다는 특징으로 인해 수리학적으로 매우 중요하다. 최근 원격탐사 기법의 발전에 따라 위성이나 드론과 같은 무인 이동체를 이용한 하천 계측법이 수질 및 지형변화 연구들에 광범위하게 적용되고 있다. 그 중 RGB 항공영상은 해상도가 높고 취득 비용이 저렴하여 확장성 및 활용도가 높다. 본 연구에서는 합류부 전단층이 촬영된 RGB 항공 영상을 이용해 합류부 전단층 분석에 활용하는 방법을 제안한다. 제안되는 방법은 RGB 항공 영상에서 본류와 지류의 수체 영역을 각각 추출하기 위해 가우시안 혼합 모형(Gaussian mixture model)을 이용한다. 추출된 수체 영역에는 자기조직화지도(self-organizing map)을 적용하고 좌표 변환을 하여 정량적인 특징을 추출한다. 본 연구에서는 알고리듬의 적용 예로서 구글어스를 통해 확보된 낙동강-남강 합류부의 항공 영상을 분석한다. 본 추출법을 이용하면 접촉식 센서를 이용하는 기존의 전단층 계측 방법들에 비해 경제적이고 안전하며 합류부 흐름의 평면적 분석을 가능하게 할 수 있을 것으로 기대된다.

• Journal of Urban Science
• /
• v.12 no.2
• /
• pp.43-52
• /
• 2023

UAV have the disadvantage that are weak from rainfall or winds due to the light platform, so use Scale-Invariant Feature Transform (SIFT) method which extrude keypoints in image matching process. To find the efficient filtering method for the construction of precise Digital Terrain Model (DTM) using UAV images, comparatively analyzed sobel and Differential of Gaussian (DoG) and found sobel is more efficient way to extrude buildings, trees, and so on. And edges are extruded more clearly when applying median additionally which have the merit of preserving edge and eliminating noise. In this study, applied sobel-median filtering which plus median to sobel and constructed the 1st filtered DTM that extrude building and trees and 2nd filtered DTM that extrude cars by threshold of gradient, Analysis of the degree of accuracy improvement showed that standard deviations of 1st filtered DTM and 2nd filtered DTM are 0.32m, 0.287m respectively, and both are acceptable for the tolerance of 0.33m for elevation points of 1/1,000 digital map, and the accuracy was increased about 10% by filtering automobiles. Plus, moving things are changed those position and direction in every image, and these are not target to filter because of the characteristic that is excluded from SIFT method.