• Journal of the Korea Society of Computer and Information
• /
• v.16 no.3
• /
• pp.211-218
• /
• 2011

Because antique Korean Peninsula maps have many historical signification, we can estimate historical evidences by analyzing them. However, it is very difficult to compare antique maps with modern maps because the antique maps were made by arranging local regions. To resolve this difficulty, we transform antique maps by rotating, scaling and translating to compare with a reference map. Each antique map is rotated in the difference of principal axis angles of the target and the reference maps, and its width and height are scaled asymmetrically using width and height ratios of bounding boxes. Finally, the two regions are overlaid by adjusting their centroids, and then the antique map is evaluated by two similarity equations. Experimental results show that the similarities of region ratio and different angle are properly computed according to era. Therefore, the proposed method can be widely used to analyze the antique Korean Peninsula maps.

• KIPS Transactions on Software and Data Engineering
• /
• v.3 no.2
• /
• pp.93-100
• /
• 2014

3D object recognition using only 2D images is a difficult work because each images are generated different to according to the view direction of cameras. Because SIFT algorithm defines the local features of the projected images, recognition result is particularly limited in case of input images with strong perspective transformation. In this paper, we propose the object recognition method that improves SIFT algorithm by using several sequential images captured from rotating 3D object around a rotation axis. We use the geometric relationship between adjacent images and merge several images into a generated feature space during recognizing object. To clarify effectiveness of the proposed algorithm, we keep constantly the camera position and illumination conditions. This method can recognize the appearance of 3D objects that previous approach can not recognize with usually SIFT algorithm.

• Korean Journal of Remote Sensing
• /
• v.16 no.1
• /
• pp.73-86
• /
• 2000

Commonly-used methods for camera modelling of pushbroom images were implemented and their performances were assessed. The models include Vector Propagation) model, Gugan and Downman(GD)'s model, Orun and Natarajan(ON)'s model, and Direct Linear Transformation(DLT) model The models were tested on a SPOT full-scene over Seoul. The number of ground control points(GCP) used range from 1 to 23. For less than 6 GCPs all other models fail except VP, with VP's accuracy being 2.7 pixels. With mode than 6 GCPs ON shows the best accuracy with 1pixel accuracy while the accuracy of VP is 1.5 pixels. GD fails in most cases due to the correlation among model parameters. The accuracy of DLT does not converge but fluctuates between 1 and 4 pixels subject to GCPs used. VP has an advantage in that its results can be used for the estimation of satellite orbit. Unresolved topics are: to remove errors in GCPs from the aforementioned accuracy value; to improve the performance of VP.

• Cross-Cultural Studies
• /
• v.37
• /
• pp.51-79
• /
• 2014

Seriously depressed by the failure in the first performance of his own drama ${\ll}$The Government Inspector${\gg}$, N. Gogol sought out a space, Italy, which is obviously a turning point for the writer. Here in Italy, the writer could be able to explore an essential foundation for the national identity as well as self-identification of Russian traditional culture, all of which have already been epitomized in the Renaissance period in Italy. The city Rome itself provided Gogol with its grandness and harmonious perfectness, influencing something 'spiritual being' upon the writer. The work under discussion, "Rome," is thus created through these literary circumstances. Though it is made under the different title as "Annuntiata" and it delivers a love story between lovers, the story lines gradually turned into a fiction about the city, Rome. In comparison with city Paris, Gogol himself presents a negative view of the French metropolitan, saying that it is nothing but a by-product of the 19th century civilization. Interestingly enough, Rome for Gogol is totally different; it is the place of sublimity, that is a locus of harmonious, holy, and eternal city. Likewise, this pattern can be said of another description on the two contradictory cities: Paris and Rome. Again, Gogol fully pictures the city Paris as centripetal and Rome as centrifugal, in which the main protagonist makes the reader indulge in his own world. Throughout the story the writer tells us a transformation experienced by his character, and the work ends with an open denouement. Like Jerusalem, Rome is the city of resurrection for Gogol. Yet, this kind of possibility of transformation in the story is exposed to the hero, and it arguably depends on the extent to which he explores the readiness for encountering of 'eternity' in this "eternal city."

• Proceedings of the Korea Information Processing Society Conference
• /
• 2010.04a
• /
• pp.510-513
• /
• 2010

Color is basically composed of hue, saturation and value. Many objects are made up with color. When people see color, they feel different emotion because of different combination of hue, saturation and value of different colors. Thus, people feel different feeling about the taste of food depending on its color. Thus, by analyzing what color makes people feel tasty about food, we can make food to look more delicious. When people take pictures of food, theyusually do not consider this into account. However if we apply this technology into taking pictures of food, we can make the food look more delicious. This technology can be applied when people want to upload pictures of food in blog, homepage and twitter and so on. In this paper, we analyze the feelings of color of people and then choose the best color combination to present food. After that we change the original image into the new one based on the analysis of color. This way, we can reflect each user's preference.

• Journal of Broadcast Engineering
• /
• v.26 no.4
• /
• pp.453-462
• /
• 2021

A lot of license plate data is required for car number recognition. License plate data needs to be balanced from past license plates to the latest license plates. However, it is difficult to obtain data from the actual past license plate to the latest ones. In order to solve this problem, a license plate recognition study through deep learning is being conducted by creating a synthetic license plates. Since the synthetic data have differences from real data, and various data augmentation techniques are used to solve these problems. Existing data augmentation simply used methods such as brightness, rotation, affine transformation, blur, and noise. In this paper, we apply a style transformation method that transforms synthetic data into real-world data styles with data augmentation methods. In addition, real license plate data are noisy when it is captured from a distance and under the dark environment. If we simply recognize characters with input data, chances of misrecognition are high. To improve character recognition, in this paper, we applied the DeblurGANv2 method as a quality improvement method for character recognition, increasing the accuracy of license plate recognition. The method of deep learning for license plate detection and license plate number recognition used YOLO-V5. To determine the performance of the synthetic license plate data, we construct a test set by collecting our own secured license plates. License plate detection without style conversion recorded 0.614 mAP. As a result of applying the style transformation, we confirm that the license plate detection performance was improved by recording 0.679mAP. In addition, the successul detection rate without image enhancement was 0.872, and the detection rate was 0.915 after image enhancement, confirming that the performance improved.

• Korean Journal of Remote Sensing
• /
• v.30 no.2
• /
• pp.275-292
• /
• 2014

The objective of this study is to determine Tasseled Cap Transformation (TCT) coefficients for the Geostationary Ocean Color Imager (GOCI). TCT is traditional method of analyzing the characteristics of the land area from multi spectral sensor data. TCT coefficients for a new sensor must be estimated individually because of different sensor characteristics of each sensor. Although the primary objective of the GOCI is for ocean color study, one half of the scene covers land area with typical land observing channels in Visible-Near InfraRed (VNIR). The GOCI has a unique capability to acquire eight scenes per day. This advantage of high temporal resolution can be utilized for detecting daily variation of land surface. The GOCI TCT offers a great potential for application in near-real time analysis and interpretation of land cover characteristics. TCT generally represents information of "Brightness", "Greenness" and "Wetness". However, in the case of the GOCI is not able to provide "Wetness" due to lack of ShortWave InfraRed (SWIR) band. To maximize the utilization of high temporal resolution, "Wetness" should be provided. In order to obtain "Wetness", the linear regression method was used to align the GOCI Principal Component Analysis (PCA) space with the MODIS TCT space. The GOCI TCT coefficients obtained by this method have different values according to observation time due to the characteristics of geostationary earth orbit. To examine these differences, the correlation between the GOCI TCT and the MODIS TCT were compared. As a result, while the GOCI TCT coefficients of "Brightness" and "Greenness" were selected at 4h, the GOCI TCT coefficient of "Wetness" was selected at 2h. To assess the adequacy of the resulting GOCI TCT coefficients, the GOCI TCT data were compared to the MODIS TCT image and several land parameters. The land cover classification of the GOCI TCT image was expressed more precisely than the MODIS TCT image. The distribution of land cover classification of the GOCI TCT space showed meaningful results. Also, "Brightness", "Greenness", and "Wetness" of the GOCI TCT data showed a relatively high correlation with Albedo ($R^2$ = 0.75), Normalized Difference Vegetation Index (NDVI) ($R^2$ = 0.97), and Normalized Difference Moisture Index (NDMI) ($R^2$ = 0.77), respectively. These results indicate the suitability of the GOCI TCT coefficients.

• Journal of IKEEE
• /
• v.26 no.2
• /
• pp.160-168
• /
• 2022

In this paper, we propose a 3D mesh reconstruction method from a single image using deep learning and a sphere shape transformation method. The proposed method has the following originality that is different from the existing method. First, the position of the vertex of the sphere is modified to be very similar to the 3D point cloud of an object through a deep learning network, unlike the existing method of building edges or faces by connecting nearby points. Because 3D point cloud is used, less memory is required and faster operation is possible because only addition operation is performed between offset value at the vertices of the sphere. Second, the 3D mesh is reconstructed by covering the surface information of the sphere on the modified vertices. Even when the distance between the points of the 3D point cloud created by correcting the position of the vertices of the sphere is not constant, it already has the face information of the sphere called face information of the sphere, which indicates whether the points are connected or not, thereby preventing simplification or loss of expression. can do. In order to evaluate the objective reliability of the proposed method, the experiment was conducted in the same way as in the comparative papers using the ShapeNet dataset, which is an open standard dataset. As a result, the IoU value of the method proposed in this paper was 0.581, and the chamfer distance value was It was calculated as 0.212. The higher the IoU value and the lower the chamfer distance value, the better the results. Therefore, the efficiency of the 3D mesh reconstruction was demonstrated compared to the methods published in other papers.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.6D
• /
• pp.1025-1032
• /
• 2006

Sensor modeling of high-resolution satellites is a prerequisite procedure for mapping and GIS applications. Sensor models, describing the geometric relationship between scene and object, are divided into two main categories, which are rigorous and approximate sensor models. A rigorous model is based on the actual geometry of the image formation process, involving internal and external characteristics of the implemented sensor. However, approximate models require neither a comprehensive understanding of imaging geometry nor the internal and external characteristics of the imaging sensor, which has gathered a great interest within photogrammetric communities. This paper described a comparison between rigorous and various approximate sensor models that have been used to determine three-dimensional positions, and proposed the appropriate sensor model in terms of the satellite imagery usage. Through the case study of using IKONOS satellite scenes, rigorous and approximate sensor models have been compared and evaluated for the positional accuracy in terms of acquirable number of ground controls. Bias compensated RFM(Rational Function Model) turned out to be the best among compared approximate sensor models, both modified parallel projection and parallel-perspective model were able to be modelled with a small number of controls. Also affine transformation, one of the approximate sensor models, can be used to determine the planimetric position of high-resolution satellites and perform image registration between scenes.

• Korean Journal of Cognitive Science
• /
• v.16 no.4
• /
• pp.305-322
• /
• 2005

Watermarking is a widely employed method tn protecting copyright of a digital image, the owner's unique image is embedded into the original image. Strengthened level of watermark insertion would help enhance its resilience in the process of extraction even from various distortions of transformation on the image size or resolution. However, its level, at the same time, should be moderated enough not to reach human visibility. Finding a balance between these two is crucial in watermarking. For the algorithm for watermarking, the predefined strength of a watermark, computed from the physical difference between the original and embedded images, is applied to all images uniformal. The mean brightness or contrast of the surrounding images, other than the absolute brightness of an object, could affect human sensitivity for object detection. In the present study, we examined whether the detectability for watermark noise might be attired by image statistics: mean brightness and contrast of the image. As the first step to examine their effect, we made rune fundamental images with varied brightness and control of the original image. For each fundamental image, detectability for watermark noise was measured. The results showed that the strength ot watermark node for detection increased as tile brightness and contrast of the fundamental image were increased. We have fitted the data to a regression line which can be used to estimate the strength of watermark of a given image with a certain brightness and contrast. Although we need to take other required factors into consideration in directly applying this formula to actual watermarking algorithm, an adaptive watermarking algorithm could be built on this formula with image statistics, such as brightness and contrast.