• International Journal of Naval Architecture and Ocean Engineering
• /
• 제13권1호
• /
• pp.187-193
• /
• 2021

In order to solve the problem of false terrains caused by environmental interferences and tunneling effect in the conventional multi-beam seafloor terrain detection, this paper proposed a seafloor topography detection method based on fast two-dimensional (2D) Censored Mean Level Detector-statistics Constant False Alarm Rate (CMLD-CFAR) method. The proposed method uses s cross-sliding window. The target occlusion phenomenon that occurs in multi-target environments can be eliminated by censoring some of the large cells of the reference cells, while the remaining reference cells are used to calculate the local threshold. The conventional 2D CMLD-CFAR methods need to estimate the background clutter power level for every pixel, thus increasing the computational burden significantly. In order to overcome this limitation, the proposed method uses a fast algorithm to select the Regions of Interest (ROI) based on a global threshold, while the rest pixels are distinguished as clutter directly. The proposed method is verified by experiments with real multi-beam data. The results show that the proposed method can effectively solve the problem of false terrain in a multi-beam terrain survey and achieve a high detection accuracy.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2021년도 춘계학술발표대회
• /
• 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.

• 대한임베디드공학회논문지
• /
• 제5권2호
• /
• pp.93-102
• /
• 2010

The FMCW(Frequency Modulation Continuous Wave) radar possesses range-velocity ambiguity to identify the correct combination of beat frequencies for each target in the multi-target situation. It can lead to ghost targets and missing targets, and it can reduce the detection probability. In this pap er, we propose an effective identification algorithm for the correct pairs of beat frequencies and the signal processing hardware architecture to effectively support the algorithm. First, using the correlation of the detected up- and down-beat frequencies and Doppler frequencies, the possible combinations are determined. Then, final pairing algorithm is completed with the power spectrum density of the correlated up- and down-beat frequencies. The proposed hardware processor has the basic architecture consisting of beat-frequency registers, pairing table memory, and decision unit. This method will be useful to improve the radar detection probability and reduce the false alarm rate.

• 말소리와 음성과학
• /
• 제1권3호
• /
• pp.95-102
• /
• 2009

In this paper, an efficient target speech detection algorithm is proposed for the performance improvement of multi-input speech enhancement. Using the normalized cross correlation value between two selected channels, the proposed algorithm estimates the probabilistic distribution function of the value from the pure noise interval. Then, log-likelihoods are calculated with the function and the normalized cross correlation value to detect the target speech interval precisely. The detection results are applied to the generalized sidelobe canceller-based algorithm. Experimental results show that the proposed algorithm significantly improves the speech recognition performance and the signal-to-noise ratios.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제17권4호
• /
• pp.1066-1079
• /
• 2023

Multi-target tracking based on the detector is a very hot and important research topic in target tracking. It mainly includes two closely related processes, namely target detection and target tracking. Where target detection is responsible for detecting the exact position of the target, while target tracking monitors the temporal and spatial changes of the target. With the improvement of the detector, the tracking performance has reached a new level. The problem that always exists in the research of target tracking is the problem that occurs again after the target is occluded during tracking. Based on this question, this paper proposes a DeepSORT model based on SIFT features to improve ship tracking. Unlike previous feature extraction networks, SIFT algorithm does not require the characteristics of pre-training learning objectives and can be used in ship tracking quickly. At the same time, we improve and test the matching method of our model to find a balance between tracking accuracy and tracking speed. Experiments show that the model can get more ideal results.

• Journal of Communications and Networks
• /
• 제15권2호
• /
• pp.164-172
• /
• 2013

We consider the problem of secure and robust clustering for quantized target tracking in wireless sensor networks (WSN) where the observed system is assumed to evolve according to a probabilistic state space model. We propose a new method for jointly activating the best group of candidate sensors that participate in data aggregation, detecting the malicious sensors and estimating the target position. Firstly, we select the appropriate group in order to balance the energy dissipation and to provide the required data of the target in the WSN. This selection is also based on the transmission power between a sensor node and a cluster head. Secondly, we detect the malicious sensor nodes based on the information relevance of their measurements. Then, we estimate the target position using quantized variational filtering (QVF) algorithm. The selection of the candidate sensors group is based on multi-criteria function, which is computed by using the predicted target position provided by the QVF algorithm, while the malicious sensor nodes detection is based on Kullback-Leibler distance between the current target position distribution and the predicted sensor observation. The performance of the proposed method is validated by simulation results in target tracking for WSN.

• 전자공학회논문지SC
• /
• 제47권5호
• /
• pp.62-68
• /
• 2010

적외선영상에서 표적을 효율적으로 탐지하는 새로운 자동표적탐지 알고리즘을 제안한다. 이 연구의 목적은 실제 야지환경에서 획득된 적외선영상에서 낮은 오경보 확률로 표적의 위치를 정확히 찾는 것이다. 제안한 방법이 기존의 방법과 다른 점은 초기 탐지단계에서 사용되는 모폴로지 필터링 기법을 밝기정보를 갖고 있는 원래 입력 영상이 아닌 가버(Gabor) 응답 영상에 적용한 것과 표적과 클러터를 구분하기 위해 표적의 정확한 윤곽선 추출을 필요로 하지않는 것이다. 제안한 방법은 크게 3단계로 구성된다. 첫째로, 영상에서 돌출된 영역을 찾기 위해 입력영상으로부터 4 방향의 가버 응답을 구하고 픽셀별로 가버응답 합 영상을 구한다. 이 영상에 모폴로지 기법을 적용하여 돌출된 영역의 위치를 찾는다. 둘째로, 원래의 입력영상의 돌출된 영역에서 지역적인 질감특징 정보들을 찾는다. 마지막 단계로, 찾아진 지역적 특징 정보들이 신경회로망인 다층퍼셉트론 (Multi-Layer Perceptron)으로 입력되어 학습된 훈련 데이터들과의 비교를 통해 실제 표적과 클러터를 구분한다. 실험에서는 제안한 방법을 군사용 적외선 영상장비를 사용하여 실제 야지 환경에 획득된 영상에 적용하여 우수성과 실용가능성을 확인한다.

• 한국전자파학회논문지
• /
• 제22권10호
• /
• pp.966-980
• /
• 2011

레이더는 표적으로부터 반사된 신호의 크기, 도플러 속도로부터 표적의 거리, 속도 정보를 알 수 있으며, 이런 반사 신호의 특징들은 표적의 반사 특성과 기동에 의해 결정된다. 표적의 위치 정보에 대한 각도 오차는 합채널에 대한 차채널의 크기 비로부터 추출된다. 본 논문에서는 다기능 레이더의 성능을 평가하고 분석하기 위한 레이더 반사 신호 모의 장치(RSSE)에 대해 소개하였다. 개발된 레이더 반사 신호 모의 장치는 다채널 직접 디지털 합성(MC-DDS)을 이용하여, 재밍 신호를 포함한 다중 표적 환경을 모사할 수 있도록 구현되었으며, 효율적인 하드웨어 구조 설계를 통해 모사할 수 있는 표적의 수를 용이하게 확장할 수 있도록 설계되었다. 개발된 모의 장치의 요구 성능 및 기능을 시험 환경에서 확인하였으며, 신호 처리기(RSP)와의 연동 시험 구성에서 표적 탐지 성능을 입증하였다.

• 한국ITS학회 논문지
• /
• 제16권2호
• /
• pp.105-115
• /
• 2017

루프 검지기, 영상 검지기 등의 기존 교통 검지기들은 설치와 유지보수에 드는 비용이 크고, 밤과 낮에 따라 상이한 검지 알고리즘이 필요하거나 날씨에 따라 검지율의 편차가 크다는 단점을 가지고 있다. 반면에 밀리미터파 레이더는 악천후에 의한 영향을 받지 않고, 주야간에 관계없이 일정한 검지 성능을 얻을 수 있다. 덧붙여 설치와 유지보수를 위하여 교통 통제의 필요가 없고, 다수의 차량을 동시에 검지 가능하다. 본 연구는 이러한 장점을 가진 레이더 센서를 활용한 다중 물체 검지 알고리즘을 기존의 단일 물체 검지 알고리즘을 응용하여 구현하였으며 이에 대한 평가를 수행하여 의미 있는 결과를 얻었다.

• 대한의용생체공학회:의공학회지
• /
• 제26권1호
• /
• pp.55-63
• /
• 2005

A novel imaging system for High-resolution Ultrasonic Transmission Tomography (HUTT) and soft tissue differentiation methodology for the HUTT system are presented. The critical innovation of the HUTT system includes the use of sub-millimeter transducer elements for both transmitter and receiver arrays and multi-band analysis of the first-arrival pulse. The first-arrival pulse is detected and extracted from the received signal (i.e., snippet) at each azimuthal and angular location of a mechanical tomographic scanner in transmission mode. Each extracted snippet is processed to yield a multi-spectral vector of attenuation values at multiple frequency bands. These vectors form a 3-D sinogram representing a multi-spectral augmentation of the conventional 2-D sinogram. A filtered backprojection algorithm is used to reconstruct a stack of multi-spectral images for each 2-D tomographic slice that allow tissue characterization. A novel methodology for soft tissue differentiation using spectral target detection is presented. The representative 2-D and 3-D HUTT images formed at various frequency bands demonstrate the high-resolution capability of the system. It is shown that spherical objects with diameter down to 0.3㎜ can be detected. In addition, the results of soft tissue differentiation and characterization demonstrate the feasibility of quantitative soft tissue analysis for possible detection of lesions or cancerous tissue.