• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.05a
• /
• pp.429-432
• /
• 2021

스마트 기기와 애플리케이션의 테스트를 위해 빠르고 정확하게 스마트 기기 화면 상에서 테스트가 필요한 위치를 추출해야 한다. 필요한 위치를 추출할 때 스마트 기기 화면과 테스트할 수 있는 영역의 매칭 방식을 사용하는데 이를 위해 이미지의 변형이 발생해도 원하는 영역의 matching point 을 빠르고 정확하게 추출하는 feature matching 방식의 D2-Net 의 feature extraction 모델과 fitting algorithm 을 변경하였다.

• The Transactions of the Korea Information Processing Society
• /
• v.13 no.7
• /
• pp.335-347
• /
• 2024

Open-vocabulary 3D point cloud instance segmentation (OV-3DIS) is a challenging visual task to segment a 3D scene point cloud into object instances of both base and novel classes. In this paper, we propose a novel model Open3DME for OV-3DIS to address important design issues and overcome limitations of the existing approaches. First, in order to improve the quality of class-agnostic 3D masks, our model makes use of T3DIS, an advanced Transformer-based 3D point cloud instance segmentation model, as mask proposal module. Second, in order to obtain semantically text-aligned visual features of each point cloud segment, our model extracts both 2D and 3D features from the point cloud and the corresponding multi-view RGB images by using pretrained CLIP and OpenSeg encoders respectively. Last, to effectively make use of both 2D and 3D visual features of each point cloud segment during label assignment, our model adopts a unique feature ensemble method. To validate our model, we conducted both quantitative and qualitative experiments on ScanNet-V2 benchmark dataset, demonstrating significant performance gains.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.24 no.1
• /
• pp.131-138
• /
• 2006

Recently, data gathering and algorithm developing are in progress for the GIS application using point feature. Several researches prove that verification of the spatial clustering and evaluation of inter-dependencies between event and control are possible. On the other hand, most of the point features as GIS data are gathered by indirect method, such as address geo-coding, rather than by direct method, such as field surveying. Futhermore, lots of statistics by administrative district based on the point features have no coordinates information of the points. In this study, calculating the K-index in GIS environment, to evaluate the effect of aggregation of raw data on K-index, K-indices estimated from raw data (parcel unit), topographically aggregated data (block unit), administratively aggregated data (administrative district unit) are compared and evaluated. As a result, point feature, highly clustered in local area, is largely distorted when aggregated administratively. But, the K-indices of topographically aggregated data is very similar to the K-indices of raw data.

• Smart Media Journal
• /
• v.6 no.4
• /
• pp.9-16
• /
• 2017

The railway point machine is an especially important component that changes the traveling direction of a train. Failure of the point machine may cause a serious railway accident. Therefore, early detection of failures is important for the management of railway condition monitoring systems. In this paper, we propose a noise-robust anomaly detection method in railway condition monitoring systems using sound data. First, we extract feature vectors from the spectrogram image of sound signals and convert it into modulation feature to ensure robust performance, and lastly, use the support vector machine (SVM) as an early anomaly detector of railway point machines. By the experimental results, we confirmed that the proposed method could detect the anomaly conditions of railway point machines with acceptable accuracy even under noisy conditions.

• Journal of KIISE:Software and Applications
• /
• v.33 no.4
• /
• pp.402-410
• /
• 2006

This paper presents a method for feature based block motion estimation that uses multi -resolution image sequences to obtain the panoramic images in the continuous ultrasound image sequences. In the conventional block motion estimation method, the centers of motion estimation blocks are set at the predetermined and equally spaced locations. This requires the large blocks to include at least one feature, which inevitably requires long estimation time. In this paper, we propose an adaptive method which locates the center of the motion estimation blocks at the feature points. This make it possible to reduce the block size while keeping the motion estimation accuracy The Harris-Stephen corner detector is used to get the feature points. The comer points tend to group together, which cause the error in the global motion estimation. In order to distribute the feature points as evenly as Possible, the image is firstly divided into regular subregions, and a strongest corner point is selected as a feature in each subregion. The ultrasound Images contain speckle patterns and noise. In order to reduce the noise artifact and reduce the computational time, the proposed method use the multi-resolution image sequences. The first algorithm estimates the motion in the smoothed low resolution image, and the estimated motion is prolongated to the next higher resolution image. By this way the size of search region can be reduced in the higher resolution image. Experiments were performed on three types of ultrasound image sequences. These were shown that the proposed method reduces both the computational time (from 77ms to 44ms) and the displaced frame difference (from 66.02 to 58.08).

• Journal of IKEEE
• /
• v.25 no.2
• /
• pp.273-279
• /
• 2021

In this paper, we propose an algorithm that shortens the learning time by performing individual learning using partitioning the deep learning structure. The proposed algorithm consists of four processes: network classification origin setting process, feature vector extraction process, feature noise removal process, and class classification process. First, in the process of setting the network classification starting point, the division starting point of the network structure for effective feature vector extraction is set. Second, in the feature vector extraction process, feature vectors are extracted without additional learning using the weights previously learned. Third, in the feature noise removal process, the extracted feature vector is received and the output value of each class is learned to remove noise from the data. Fourth, in the class classification process, the noise-removed feature vector is input to the multi-layer perceptron structure, and the result is output and learned. To evaluate the performance of the proposed algorithm, we experimented with the Extended Yale B face database. As a result of the experiment, in the case of the time required for one-time learning, the proposed algorithm reduced 40.7% based on the existing algorithm. In addition, the number of learning up to the target recognition rate was shortened compared with the existing algorithm. Through the experimental results, it was confirmed that the one-time learning time and the total learning time were reduced and improved over the existing algorithm.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.6
• /
• pp.99-104
• /
• 2021

The technology of estimating a movement trajectory using a monocular camera such as a smartphone and composing a surrounding 3D image is key not only in indoor positioning but also in the metaverse service. The most important thing in this technique is to estimate the coordinates of the moving camera center. In this paper, a new algorithm for geometrically estimating the moving distance is proposed. The coordinates of the 3D object point are obtained from the first and second photos, and the movement distance vector is obtained using the matching feature points of the first and third photos. Then, while moving the coordinates of the origin of the third camera, a position where the 3D object point and the feature point of the third picture coincide is obtained. Its possibility and accuracy were verified by applying it to actual continuous image data.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.8 no.2
• /
• pp.36-44
• /
• 2009

In this paper, we suggest a system of detecting a vehicle with lane violation, which can detect the vehicle with lane violation, by using the feature point tracking. The whole algorithm in the suggested system of detecting a vehicle with lane violation is composed of three stages such as feature extraction, register and tracking in feature for the tracking-targeted vehicle, and detecting a vehicle with lane violation. In the stage of feature extraction, the feature is extracted from the inputted image by sing the feature-extraction algorithm available for the real-time processing. The extracted features are again selected the racking-targeted feature. The registered feature is tracked by using NCC(normalized cross correlation). Finally, whether or not lane violation is finally detected by using information on the tracked features. As a result of experimenting the suggested system by using the acquired image in the section with a ban on intervention, the excellent performance was shown with 99.09% for positive recognition ratio and 0.9% for error ratio. The fast processing speed could be obtained in 34.48 frames per second available for real-time processing.

• Journal of Institute of Control, Robotics and Systems
• /
• v.22 no.10
• /
• pp.833-838
• /
• 2016

This paper presents a performance evaluation of various feature-initialization algorithms for forward-viewing mono-camera based simultaneous localization and mapping (SLAM), specifically in indoor environments. For mono-camera based SLAM, the position of feature points cannot be known from a single view; therefore, it should be estimated from a feature initialization method using multiple viewpoint measurements. The accuracy of the feature initialization method directly affects the accuracy of the SLAM system. In this study, four different feature initialization algorithms are evaluated in simulations, including linear triangulation; depth parameterized, linear triangulation; weighted nearest point triangulation; and particle filter based depth estimation algorithms. In the simulation, the virtual feature positions are estimated when the virtual robot, containing a virtual forward-viewing mono-camera, moves forward. The results show that the linear triangulation method provides the best results in terms of feature-position estimation accuracy and computational speed.

• Journal of Information Processing Systems
• /
• v.16 no.3
• /
• pp.688-698
• /
• 2020

Point clouds have ability to express the spatial entities, however, the point clouds redundancy always involves some uncertainties in computer recognition and model construction. Therefore, point clouds thinning is an indispensable step in point clouds model reconstruction and other applications. To overcome the shortcomings of complex classification index and long time consuming in existing point clouds thinning algorithms, this paper proposes a point clouds fast thinning algorithm. Specifically, the two-dimensional index is established in plane linear array (x, y) for the scanned point clouds, and the thresholds of adjacent point distance difference and height difference are employed to further delete or retain the selected sample point. Sequentially, the index of sample point is traversed forwardly and backwardly until the process of point clouds thinning is completed. The results suggest that the proposed new algorithm can be applied to different targets when the thresholds are built in advance. Besides, the new method also performs superiority in time consuming, modelling accuracy and feature retention by comparing with octree thinning algorithm.