• The Bulletin of The Korean Astronomical Society
• /
• v.44 no.1
• /
• pp.54.2-54.2
• /
• 2019

Removing noise which occurs inevitably when taking image data has been a big concern. There is a way to raise signal-to-noise ratio and it is regarded as the only way, image stacking. Image stacking is averaging or just adding all pixel values of multiple pictures taken of a specific area. Its performance and reliability are unquestioned, but its weaknesses are also evident. Object with fast proper motion can be vanished, and most of all, it takes too long time. So if we can handle single shot image well and achieve similar performance, we can overcome those weaknesses. Recent developments in deep learning have enabled things that were not possible with former algorithm-based programming. One of the things is generating data with more information from data with less information. As a part of that, we reproduced stacked image from single shot image using a kind of deep learning, conditional generative adversarial network (cGAN). r-band camcol2 south data were used from SDSS Stripe 82 data. From all fields, image data which is stacked with only 22 individual images and, as a pair of stacked image, single pass data which were included in all stacked image were used. All used fields are cut in $128{\times}128$ pixel size, so total number of image is 17930. 14234 pairs of all images were used for training cGAN and 3696 pairs were used for verify the result. As a result, RMS error of pixel values between generated data from the best condition and target data were $7.67{\times}10^{-4}$ compared to original input data, $1.24{\times}10^{-3}$. We also applied to a few test galaxy images and generated images were similar to stacked images qualitatively compared to other de-noising methods. In addition, with photometry, The number count of stacked-cGAN matched sources is larger than that of single pass-stacked one, especially for fainter objects. Also, magnitude completeness became better in fainter objects. With this work, it is possible to observe reliably 1 magnitude fainter object.

• Journal of Information Processing Systems
• /
• v.12 no.2
• /
• pp.183-195
• /
• 2016

This paper presents a complete method for vehicle detection and tracking in a fixed setting based on computer vision. Vehicle detection is performed based on Scale Invariant Feature Transform (SIFT) feature matching. With SIFT feature detection and matching, the geometrical relations between the two images is estimated. Then, the previous image is aligned with the current image so that moving vehicles can be detected by analyzing the difference image of the two aligned images. Vehicle tracking is also performed based on SIFT feature matching. For the decreasing of time consumption and maintaining higher tracking accuracy, the detected candidate vehicle in the current image is matched with the vehicle sample in the tracking sample set, which contains all of the detected vehicles in previous images. Most remarkably, the management of vehicle entries and exits is realized based on SIFT feature matching with an efficient update mechanism of the tracking sample set. This entire method is proposed for highway traffic environment where there are no non-automotive vehicles or pedestrians, as these would interfere with the results.

• IEIE Transactions on Smart Processing and Computing
• /
• v.4 no.5
• /
• pp.311-317
• /
• 2015

The aim of this study is to develop a 3D registration algorithm for positron emission tomography/computed tomography (PET/CT) and magnetic resonance (MR) images acquired from independent PET/CT and MR imaging systems. Combined PET/CT images provide anatomic and functional information, and MR images have high resolution for soft tissue. With the registration technique, the strengths of each modality image can be combined to achieve higher performance in diagnosis and radiotherapy planning. The proposed method consists of two stages: normalized mutual information (NMI)-based global matching and independent component analysis (ICA)-based refinement. In global matching, the field of view of the CT and MR images are adjusted to the same size in the preprocessing step. Then, the target image is geometrically transformed, and the similarities between the two images are measured with NMI. The optimization step updates the transformation parameters to efficiently find the best matched parameter set. In the refinement stage, ICA planes from the windowed image slices are extracted and the similarity between the images is measured to determine the transformation parameters of the control points. B-spline. based freeform deformation is performed for the geometric transformation. The results show good agreement between PET/CT and MR images.

• ETRI Journal
• /
• v.34 no.1
• /
• pp.87-97
• /
• 2012

This study presents an automatic matching method for generating a dense, accurate, and discontinuity-preserved digital surface model (DSM) using multiple images acquired by an aerial digital frame camera. The proposed method consists of two main procedures: area-based multi-image matching (AMIM) and stereo-pair epipolar line matching (SELM). AMIM evaluates the sum of the normalized cross correlation of corresponding image points from multiple images to determine the optimal height of an object point. A novel method is introduced for determining the search height range and incremental height, which are necessary for the vertical line locus used in the AMIM. This procedure also includes the means to select the best reference and target images for each strip so that multi-image matching can resolve the common problem over occlusion areas. The SELM extracts densely positioned distinct points along epipolar lines from the multiple images and generates a discontinuity-preserved DSM using geometric and radiometric constraints. The matched points derived by the AMIM are used as anchor points between overlapped images to find conjugate distinct points using epipolar geometry. The performance of the proposed method was evaluated for several different test areas, including urban areas.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.25 no.6_1
• /
• pp.529-536
• /
• 2007

Registration between image and object space is a fundamental step in photogrammetry and computer vision. Along with rapid development of sensors - multi/hyper spectral sensor, laser scanning sensor, radar sensor etc., the needs for registration between different sensors are ever increasing. There are two important considerations on different sensor registration. They are sensor invariant feature extraction and correspondence between them. Since point to point correspondence does not exist in image and laser scanning data, it is necessary to have higher entities for extraction and correspondence. This leads to modify first, existing mathematical and geometrical model which was suitable for point measurement to line measurements, second, matching scheme. In this research, linear feature is selected for sensor invariant features and matching entity. Linear features are incorporated into mathematical equation in the form of extended collinearity equation for registration problem known as photo resection which calculates exterior orientation parameters. The other emphasis is on the scheme of finding matched entities in the aide of RANSAC (RANdom SAmple Consensus) in the absence of correspondences. To relieve computational load which is a common problem in sampling theorem, deterministic sampling technique and selecting 4 line features from 4 sectors are applied.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.28 no.3
• /
• pp.385-392
• /
• 2010

A LiDAR technique has the merits that survey engineers can get a large number of measurements with high precision quickly. Aerial photos and satellite sensor images are used for generating 3D spatial images which are matched with the map coordinates and elevation data from digital topographic files. Also, those images are used for matching with 3D spatial image contents through perspective view condition composed along to the designated roads until arrival the corresponding location. Recently, 3D aviation image could be generated by various digital data. The advanced geographical methods for guidance of the destination road are experimented under the GIS environments. More information and access designated are guided by the multimedia contents on internet or from the public tour information desk using the simulation images. The height data based on LiDAR is transformed into DEM, and the real time unification of the vector via digital image mapping and raster via extract evaluation are transformed to trace the generated model of 3-dimensional downtown building along to the long distance for 3D tract model generation.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.22 no.7
• /
• pp.40-47
• /
• 2014

This paper proposes a matching method of GIS lane information with a vehicle position using camera image to reduce DGPS error. Images of straight road are taken using a camera that is installed on the front center of the vehicle, and the distance between the vehicle and the lane are estimated using the images. The current GIS lane data is matched by comparing the estimated distance and the measured distance using a DGPS. Inverse perspective mapping is used to minimize the error of image processing from the heading angle, and single buffering method is applied to decide the exact moment of GIS match. Through practical test on the highway, feasibility of the GIS matching using camera image is confirmed.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.20 no.3
• /
• pp.331-341
• /
• 2002

In order to extract 3-dimensional information from 2-D image, stereo images are prerequisite. Moreover, for the measurement of moving objects, the synchronized sequential stereo images have to be captured and image matching should be implemented for determining the location of moving objects. In this research, a simple method computing 3-dimensional coordinates from sequential images of moving objects was implemented. The sequential stereo images were captured by a video camera with a beam splitter. Once video images were digitalized by frame grabber, the interest points were extracted and matched in each stereo image, and the coordinates of center of them are calculated using weighted average method. Then, 3-dimensional coordinates of moving objects were computed by DLT algorithms.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.8 no.4
• /
• pp.14-23
• /
• 2005

This paper proposes a method for matching two different images, especially satellite images. In the general image matching fields, when an image is compared to other image, they may have different properties on the size, contents, brightness, etc. If there is no noise in each image, in other words, they have identical pixel level and unchanged edges, the image matching method will be simple comparison between two images with pixel by pixel. However, in many applications, most of images to be matched should have much different properties. This paper proposes an efficient method for matching satellite images. This method is to match a raw satellite image with GCP chips. From this we can make a geometrically corrected image. The proposed method is based on wavelet transformation, not required any pre-processing such as histogram equalization, analysis of raw image like the traditional methods.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.31 no.2
• /
• pp.165-172
• /
• 2013

In this paper, high-quality mosaic image is generated by high-resolution hyperspectral strip images using scale-invariant feature transform (SIFT) algorithm, which is one of the representative image matching methods. The experiments are applied to AISA Eagle images geo-referenced by using GPS/INS information acquired when it was taken on flight. The matching points between three strips of hyperspectral images are extracted using SIFT method, and the transformation models between images are constructed from the points. Mosaic image is, then, generated using the transformation models constructed from corresponding images. Optimal band appropriate for the matching point extraction is determined by selecting representative bands of hyperspectral data and analyzing the matched results based on each band. Mosaic image generated by proposed method is visually compared with the mosaic image generated from initial geo-referenced AISA hyperspectral images. From the comparison, we could estimate geometrical accuracy of generated mosaic image and analyze the efficiency of our methodology.