• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제14권4호
• /
• pp.249-255
• /
• 2014

We present a simple method to watermark three-dimensional (3D) triangular meshes that have been generated from the depth data of the Kinect sensor. In contrast to previous methods, which maintain the shape of 3D triangular meshes and decide the embedding place, requiring calculations of vertices and their neighbors, our method is based on selecting one of the coordinate axes. To maintain shape, we use discrete wavelet transform and constant regularization. We know that the watermarking system needs the information to be embedded; we used a text to provide that information. We used geometry attacks such as rotation, scales, and translation, to test the performance of this watermarking system. Performance parameters in this paper include the vertices error rate (VER) and bit error rate (BER). The results from the VER and BER indicate that using a correction term before the extraction process makes our system robust to geometry attacks.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2002년도 ITC-CSCC -3
• /
• pp.1678-1680
• /
• 2002

Stereo matching technique is used in many practical fields like satellite image analysis and computer vision. In this paper, we suggest a method to extract a target object image from a complicated background. For example, human face image can be extracted from random background. This method can be applied to computer vision such as security system, dressing simulation by use of extracted human face, 3D modeling, and security system. Many researches about stereo matching have been performed. Conventional approaches can be categorized into area-based and feature-based method. In this paper, we start from area-based method and apply area tracking using scanning window. Coarse depth information is used for area merging process using area searching data. Finally, we produce a target object image.

• 한국통신학회논문지
• /
• 제15권12호
• /
• pp.1002-1016
• /
• 1990

본 논문은 스테레오비젼 시스템을 설계 제작하고, 제작된 시스템을 가지고 3차원 이동물체의 변위평가 방법을 제안하였다. 이동물체의 추출은 차영상 알고리즘에 의해 추출하고, 3차원 이동물체의 기하학적인 위치좌표는 2개의 2차원 물체의 면적중심을 합성시켜 구하였다. 3차원 이동물체의 범위평가는 합성된 3차원 좌표값에 의해서 물체의 이동속도 및 거리, 이동궤적, 카메라와 물체 사이의 공간거리를 산출하여 입증하였다.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2012년도 한국컴퓨터종합학술대회논문집 Vol.39 No.1(C)
• /
• pp.448-450
• /
• 2012

본 논문에서는 Kinect 방식의 RGB-D 영상센서를 사용하여, 깊이(Depth) 영상으로부터 3차원 객체의 기하정보를 표현하는 표면 정규 벡터(Surface Normal Vector)를 추출하고, 그 결과를 영상화하는 방법을 제안하며, 제안된 방법으로 생성된 영상으로부터 깊이 영상의 특징점 및 특징 기술자를 추출하여 3차원 객체 인식 성능을 향상시키는 방법을 제안한다. 또한 생성된 RGB-D 특징 기술자들을 객체 단위로 구분 가능한 코드북(CodeBook) 학습을 통한 인식방법을 제안하여 객체의 인식 성능을 높이는 방법을 제안한다. 제안하는 RGB-D 기반의 특징 추출 및 학습 방법은 텍스쳐 유무, 카메라 회전 및 이동 변화 등의 환경변화에 강건함을 실험적으로 증명하였으며, 이 방법은 Kinect 방식의 RGB-D 영상을 사용하는 3차원 객체/공간 인식 및 추적, 혹은 이를 응용하는 증강현실 시스템에 적용하여 사용될 수 있다.

• 한국멀티미디어학회논문지
• /
• 제24권5호
• /
• pp.695-700
• /
• 2021

The Faster RCNN-based shellfish recognition algorithm is introduced for shellfish recognition studies that currently do not have any deep learning-based algorithms in a practical setting. The original feature extraction module is replaced by DenseNet, which fuses multi-level feature data and optimises the NMS algorithm, network depth and merging method; overcoming the omission of shellfish overlap, multiple shellfish and insufficient light, effectively solving the problem of low shellfish classification accuracy. In the complexifier test environment, the test accuracy was improved by nearly 4%. Higher testing accuracy was achieved compared to the original testing algorithm. This provides favourable technical support for future applications of the improved Faster RCNN approach to seafood quality classification.

• 한국멀티미디어학회논문지
• /
• 제25권6호
• /
• pp.807-818
• /
• 2022

In order to further improve the accuracy and time efficiency of behavior recognition in intelligent monitoring scenarios, a human behavior recognition algorithm based on YOLO combined with LSTM and CNN is proposed. Using the real-time nature of YOLO target detection, firstly, the specific behavior in the surveillance video is detected in real time, and the depth feature extraction is performed after obtaining the target size, location and other information; Then, remove noise data from irrelevant areas in the image; Finally, combined with LSTM modeling and processing time series, the final behavior discrimination is made for the behavior action sequence in the surveillance video. Experiments in the MSR and KTH datasets show that the average recognition rate of each behavior reaches 98.42% and 96.6%, and the average recognition speed reaches 210ms and 220ms. The method in this paper has a good effect on the intelligence behavior recognition.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권7호
• /
• pp.2407-2424
• /
• 2022

Recently, Transformer has made great progress in video retrieval tasks due to its high representation capability. For the structure of a Transformer, the cascaded self-attention modules are capable of capturing long-distance feature dependencies. However, the local feature details are likely to have deteriorated. In addition, increasing the depth of the structure is likely to produce learning bias in the learned features. In this paper, an improved Transformer structure named TransDCS (Transformer with Dynamic Convolution and Shortcut) is proposed. A Multi-head Conv-Self-Attention module is introduced to model the local dependencies and improve the efficiency of local features extraction. Meanwhile, the augmented shortcuts module based on a dual identity matrix is applied to enhance the conduction of input features, and mitigate the learning bias. The proposed model is tested on MSRVTT, LSMDC and Activity-Net benchmarks, and it surpasses all previous solutions for the video-text retrieval task. For example, on the LSMDC benchmark, a gain of about 2.3% MdR and 6.1% MnR is obtained over recently proposed multimodal-based methods.

• 한국정밀공학회지
• /
• 제10권4호
• /
• pp.170-179
• /
• 1993

In this study, self-organized neural network is used to solve the vorrespondence problem of the axial stereo image. Edge points are extracted from a pair of stereo images and then the edge points of rear image are assined to the output nodes of neural network. In the matching process, the two input nodes of neural networks are supplied with the coordi- nates of the edge point selected randomly from the front image. This input data activate optimal output node and its neighbor nodes whose coordinates are thought to be correspondence point for the present input data, and then their weights are allowed to updated. After several iterations of updating, the weights whose coordinates represent rear edge point are converged to the coordinates of the correspondence points in the front image. Because of the feature map properties of self-organized neural network, noise-free and smoothed depth data can be achieved.

• Ocean Systems Engineering
• /
• 제13권3호
• /
• pp.247-268
• /
• 2023

Tidal stream energy extraction remains a site-specific resource due to the "first generation" criteria requiring high-velocity tidal streams. Most studies on tidal energy and turbine blade design heavily focus on installation sites with higher velocity conditions that are non-existent in tropical countries such as the Philippines. To shorten this gap, this review paper tackles tidal turbine design considerations for low-energetic regions such as the tropics. In-depth discussions of operating principles, methods of analysis, and designs of tidal turbine blades are presented. Notable tidal stream projects around the world are also mentioned in the paper. Also, it provides a perspective on the potential of this renewable energy to produce electricity for various sites in the Philippines. Finally, the paper emphasizes the need for new tidal turbine blade designs to be viable in tropical regions, such as the Philippines.

• Communications for Statistical Applications and Methods
• /
• 제30권6호
• /
• pp.551-560
• /
• 2023

Recently, there has been significant research into the recognition of human activities using three-dimensional sequential skeleton data captured by the Kinect depth sensor. Many of these studies employ deep learning models. This study introduces a novel feature selection method for this data and analyzes it using machine learning models. Due to the high-dimensional nature of the original Kinect data, effective feature extraction methods are required to address the classification challenge. In this research, we propose using the first four moments as predictors to represent the distribution of joint sequences and evaluate their effectiveness using two datasets: The exergame dataset, consisting of three activities, and the MSR daily activity dataset, composed of ten activities. The results show that the accuracy of our approach outperforms existing methods on average across different classifiers.