• International Journal of Automotive Technology
• /
• v.7 no.3
• /
• pp.361-367
• /
• 2006

This paper describes a novel stereo vision based localization of free parking site, which recognizes the target position of automatic parking system. Pixel structure classification and feature based stereo matching extract the 3D information of parking site in real time. The pixel structure represents intensity configuration around a pixel and the feature based stereo matching uses step-by-step investigation strategy to reduce computational load. This paper considers only parking site divided by marking, which is generally drawn according to relevant standards. Parking site marking is separated by plane surface constraint and is transformed into bird's eye view, on which template matching is performed to determine the location of parking site. Obstacle depth map, which is generated from the disparity of adjacent vehicles, can be used as the guideline of template matching by limiting search range and orientation. Proposed method using both the obstacle depth map and the bird's eye view of parking site marking increases operation speed and robustness to visual noise by effectively limiting search range.

• Journal of the Korean Society of Industry Convergence
• /
• v.24 no.6_1
• /
• pp.693-698
• /
• 2021

As technology advances, there is increasing need for research in different fields where this technology is applied. On of the most researched topic in computer vision is Human activity recognition (HAR), which has widely been implemented in various fields which include healthcare, video surveillance and education. We therefore present in this paper a human activity recognition system based on scale and rotation while employing the Kinect depth sensors to obtain the human skeleton joints. In contrast to previous approaches that use joint angles, in this paper we propose that each limb has an angle with the X, Y, Z axes which we employ as feature vectors. The use of the joint angles makes our system scale invariant. We further calculate the body relative direction in the egocentric coordinates in order to provide the rotation invariance. For the system parameters, we employ 8 limbs with their corresponding angles each having the X, Y, Z axes from the coordinate system as feature vectors. The extracted features are finally trained and tested with the Long short term memory (LSTM) Network which gives us an average accuracy of 98.3%.

• ETRI Journal
• /
• v.45 no.6
• /
• pp.1022-1034
• /
• 2023

In this study, the impact of varying lighting conditions on recognition and decision-making was considered. The luminosity approach was presented to increase gesture recognition performance under varied lighting. An efficient framework was proposed for sensor-based sign language gesture identification, including picture acquisition, preparing data, obtaining features, and recognition. The depth images were collected using multiple Microsoft Kinect devices, and data were acquired by varying resolutions to demonstrate the idea. A case study was designed to attain acceptable accuracy in gesture recognition under variant lighting. Using American Sign Language (ASL), the dataset was created and analyzed under various lighting conditions. In ASL-based images, significant feature points were selected using the scale-invariant feature transformation (SIFT). Finally, an artificial neural network (ANN) classified hand gestures using specified characteristics for validation. The suggested method was successful across a variety of illumination conditions and different image sizes. The total effectiveness of NN architecture was shown by the 97.6% recognition accuracy rate of 26 alphabets dataset with just a 2.4% error rate.

• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.2
• /
• pp.61-66
• /
• 2019

Augmented reality is a technology that makes a virtual object appear as if it exists in reality by composing a virtual object in real time with the image captured by the camera. In order to augment the virtual object on the object existing in reality, the homography of images utilized to estimate the position and orientation of the object. The homography can be estimated by applying the RANSAC algorithm to the feature points of the images. But the homography estimation method using the RANSAC algorithm has a problem that accurate homography can not be estimated when there are many feature points in the background. In this paper, we propose a method to filter feature points of a background when the object is near and the background is relatively far away. First, we classified the depth image into relatively near region and a distant region using the Otsu's method and improve homography estimation performance by filtering feature points on the relatively distant area. As a result of experiment, processing time is shortened 71.7% compared to a conventional homography estimation method, and the number of iterations of the RANSAC algorithm was reduced 69.4%, and Inlier rate was increased 16.9%.

• Journal of Korea Society of Industrial Information Systems
• /
• v.23 no.2
• /
• pp.1-9
• /
• 2018

In this paper, we propose a face recognition method which is not sensitive to illumination change and prevents false recognition of photographs. The proposed method uses infrared and depth images at the same time, solves sensitivity of illumination change by infrared image, and prevents false recognition of two - dimensional image such as photograph by depth image. Face detection method using infrared and depth images simultaneously and feature extraction and matching method for face recognition are realized. Simulation results show that accuracy of face recognition is increased compared to conventional methods.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.9
• /
• pp.3924-3940
• /
• 2020

Dynamic hand gesture recognition(d-HGR) plays an important role in human-computer interaction(HCI) system. With the growth of hand-pose estimation as well as 3D depth sensors, depth, and the hand-skeleton dataset is proposed to bring much research in depth and 3D hand skeleton approaches. However, it is still a challenging problem due to the low resolution, higher complexity, and self-occlusion. In this paper, we propose a hand-shape feature extraction(HSFE) network to produce robust hand-shapes. We build a hand-shape model, and hand-skeleton based on LSTM to exploit the temporal information from hand-shape and motion changes. Fusion between two models brings the best accuracy in dynamic hand gesture (DHG) dataset.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.2
• /
• pp.763-774
• /
• 2015

Hand posture recognition has been a wide region of applications in Human Computer Interaction and Computer Vision for many years. The problem arises mainly due to the high dexterity of hand and self-occlusions created in the limited view of the camera or illumination variations. To remedy these problems, a hand posture recognition method using 3-D point cloud is proposed to explicitly utilize 3-D information from depth maps in this paper. Firstly, hand region is segmented by a set of depth threshold. Next, hand image normalization will be performed to ensure that the extracted feature descriptors are scale and rotation invariant. By robustly coding and pooling 3-D facets, the proposed descriptor can effectively represent the various hand postures. After that, SVM with Gaussian kernel function is used to address the issue of posture recognition. Experimental results based on posture dataset captured by Kinect sensor (from 1 to 10) demonstrate the effectiveness of the proposed approach and the average recognition rate of our method is over 96%.

• IEIE Transactions on Smart Processing and Computing
• /
• v.1 no.1
• /
• pp.1-7
• /
• 2012

This paper presents a system used to automatically recognize the road traffic control gestures of police officers. In this approach,the control gestures of traffic police officers are captured in the form of depth images.A human skeleton is then constructed using a kinematic model. The feature vector describing a traffic control gesture is built from the relative angles found amongst the joints of the constructed human skeleton. We utilize Support Vector Machines (SVMs) to perform the gesture recognition. Experiments show that our proposed method is robust and efficient and is suitable for real-time application. We also present a testbed system based on the SVMs trained data for real-time traffic gesture recognition.

• Journal of Sensor Science and Technology
• /
• v.15 no.5
• /
• pp.328-333
• /
• 2006

Magnetic Flux Leakage (MFL) methods are widely employed for the nondestructive evaluation (NDE) of gas pipelines. In the application of MFL pipeline inspection technology, corrosion anomalies are detected and identified via their leakage filed due to changes in wall thickness. The gas industry is keenly interested in automating the interpretation process, because a large amount of data to be analyzed is generated for in-line inspection. This paper presents a novel approach to the tasks of data segmentation, feature extraction and depth estimation from gas pipelines. Also, we will show that the proposed method successfully identifying artificial defects.

• Nuclear Engineering and Technology
• /
• v.52 no.7
• /
• pp.1452-1461
• /
• 2020

A triple bubbler sensor was tested in LiCl-KCl molten salt from 450 to 525 ℃ in a transparent furnace to validate thermal-expansion corrections and provide additional molten salt data sets for calibration and validation of the sensor. In addition to these tests, a model was identified and further developed to accurately determine the density, surface tension, and depth from the measured bubble pressures. A unique feature of the model is that calibration constants can be estimated using independent depth measurements, which allow calibration and validation of the sensor in an electrorefiner where the salt density and surface tension are largely unknown. This model and approach were tested using the current and previous triple bubbler data sets, and results indicate that accuracies are as high as 0.03%, 4.6%, and 0.15% for density, surface tension, and depth, respectively.