• Proceedings of the KIEE Conference
• /
• 2005.10b
• /
• pp.6-8
• /
• 2005

Study that uses geometrical information in computer vision is lively. Problem that should be preceded is matching problem before studying. Feature point should be extracted for well matching. There are a lot of methods that extract feature point from former days are studied. Because problem does not exist algorithm that is applied for all images, it is a hot water. Specially, it is not easy to find feature point in endoscope image. The big problem can not decide easily a point that is predicted feature point as can know even if see endoscope image as eyes. Also, accuracy of matching problem can be decided after number of feature points is enough and also distributed on whole image. In this paper studied algorithm that can apply to endoscope image. SIFT method displayed excellent performance when compared with alternative way (Affine invariant point detector etc.) in general image but SIFT parameter that used in general image can't apply to endoscope image. The gual of this paper is abstraction of feature point on endoscope image that controlled by contrast threshold and curvature threshold among the parameters for applying SIFT method on endoscope image. Studied about method that feature points can have good distribution and control number of feature point than traditional alternative way by controlling the parameters on experiment result.

• Journal of Broadcast Engineering
• /
• v.23 no.4
• /
• pp.559-573
• /
• 2018

Recently, one of the most actively applied image media in the most fields such as virtual reality (VR) is omni-directional or panorama image. This image is generated by stitching images obtained by various methods. In this process, it takes the most time to extract keypoints necessary for stitching. In this paper, we analyze the parameters involved in the extraction of SIFT keypoints with the aim of reducing the computation time for extracting the most widely used SIFT keypoints. The parameters considered in this paper are the initial standard deviation of the Gaussian kernel used for Gaussian filtering, the number of gaussian difference image sets for extracting local extrema, and the number of octaves. As the SIFT algorithm, the Lowe scheme, the originally proposed one, and the Hess scheme which is a convolution cascade scheme, are considered. First, the effect of each parameter value on the computation time is analyzed, and the effect of each parameter on the stitching performance is analyzed by performing actual stitching experiments. Finally, based on the results of the two analyses, we extract parameter value set that minimize computation time without degrading.

• Journal of the HCI Society of Korea
• /
• v.2 no.2
• /
• pp.35-43
• /
• 2007

This paper presents a system for image matching and recognition based on image feature detection and description techniques from artificial satellite photographs. We propose some kind of parameters from the varied environmental elements happen by image handling process. The essential point of this experiment is analyzes that affects match rate and recognition accuracy when to change of state of each parameter. The proposed system is basically inspired by Lowe's SIFT(Scale-Invariant Transform Feature) algorithm. The descriptors extracted from local affine invariant regions are saved into database, which are defined by k-means performed on the 128-dimensional descriptor vectors on an artificial satellite photographs from Google earth. And then, a label is attached to each cluster of the feature database and acts as guidance for an appeared building's information in the scene from camera. This experiment shows the various parameters and compares the affected results by changing parameters for the process of image matching and recognition. Finally, the implementation and the experimental results for several requests are shown.

• Korean Journal of Remote Sensing
• /
• v.26 no.1
• /
• pp.39-45
• /
• 2010

This study used the keypoint observed simultaneously on two images and on twodimensional intensity image data, which was obtained along with the two point clouds data that were approached for automatic focus among points on terrestrial LiDAR data, and selected matching point through SIFT algorithm. Also, for matching error diploid, RANSAC algorithm was applied to improve the accuracy of focus. As calculating the degree of three-dimensional rotating transformation, which is the transformation-type parameters between two points, and also the moving amounts of vertical/horizontal, the result was compared with the existing result by hand. As testing the building of College of Science at Konkuk University, the difference of the transformation parameters between the one through automatic matching and the one by hand showed 0.011m, 0.008m, and 0.052m in X, Y, Z directions, which concluded to be used as the data for automatic focus.

• Journal of Internet Computing and Services
• /
• v.20 no.4
• /
• pp.91-102
• /
• 2019

In autonomous driving systems, the ability to classify pedestrians in images captured by cameras is very important for pedestrian safety. In the past, after extracting features of pedestrians with HOG(Histogram of Oriented Gradients) or SIFT(Scale-Invariant Feature Transform), people classified them using SVM(Support Vector Machine). However, extracting pedestrian characteristics in such a handcrafted manner has many limitations. Therefore, this paper proposes a method to classify pedestrians reliably and effectively using CNN's(Convolutional Neural Network) deep features and transfer learning. We have experimented with both the fixed feature extractor and the fine-tuning methods, which are two representative transfer learning techniques. Particularly, in the fine-tuning method, we have added a new scheme, called M-Fine(Modified Fine-tuning), which divideslayers into transferred parts and non-transferred parts in three different sizes, and adjusts weights only for layers belonging to non-transferred parts. Experiments on INRIA Person data set with five CNN models(VGGNet, DenseNet, Inception V3, Xception, and MobileNet) showed that CNN's deep features perform better than handcrafted features such as HOG and SIFT, and that the accuracy of Xception (threshold = 0.5) isthe highest at 99.61%. MobileNet, which achieved similar performance to Xception and learned 80% fewer parameters, was the best in terms of efficiency. Among the three transfer learning schemes tested above, the performance of the fine-tuning method was the best. The performance of the M-Fine method was comparable to or slightly lower than that of the fine-tuningmethod, but higher than that of the fixed feature extractor method.

• Journal of the Institute of Convergence Signal Processing
• /
• v.12 no.1
• /
• pp.26-32
• /
• 2011

This paper proposes a new method to process panoramic image stitching using SURF(Speeded Up Robust Features). Panoramic image stitching is considered a problem of the correspondence matching. In computer vision, it is difficult to find corresponding points in variable environment where a scale, rotation, view point and illumination are changed. However, SURF algorithm have been widely used to solve the problem of the correspondence matching because it is faster than SIFT(Scale Invariant Feature Transform). In this work, we also describe an efficient approach to decreasing computation time through the homography estimation using RANSAC(random sample consensus). RANSAC is a robust estimation procedure that uses a minimal set of randomly sampled correspondences to estimate image transformation parameters. Experimental results show that our method is robust to rotation, zoom, Gaussian noise and illumination change of the input images and computation time is greatly reduced.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.6
• /
• pp.2230-2250
• /
• 2015

Multiple projectors with partially overlapping regions can be used to project a seamless image on a large projection surface. With the advent of high-resolution photography, such systems are gaining popularity. Experts set up such projection systems by subjectively identifying the types of errors induced by the system in the projected images and rectifying them by optimizing (correcting) the parameters associated with the system. This requires substantial time and effort, thus making it difficult to set up such systems. Moreover, comparing the performance of different multi-projector display (MPD) systems becomes difficult because of the subjective nature of evaluation. In this work, we present a framework to quantitatively determine the quality of an MPD system and any image projected using such a system. We have divided the quality assessment into geometric and photometric qualities. For geometric quality assessment, we use Feature Similarity Index (FSIM) and distance-based Scale Invariant Feature Transform (SIFT). For photometric quality assessment, we propose to use a measure incorporating Spectral Angle Mapper (SAM), Intensity Magnitude Ratio (IMR) and Perceptual Color Difference (ΔE). We have tested the proposed framework and demonstrated that it provides an acceptable method for both quantitative evaluation of MPD systems and estimation of the perceptual quality of any image projected by them.