• ETRI Journal
• /
• v.42 no.3
• /
• pp.399-410
• /
• 2020

Detecting visual relationships in an image is important in an image understanding task. It enables higher image understanding tasks, that is, predicting the next scene and understanding what occurs in an image. A visual relationship comprises of a subject, a predicate, and an object, and is related to visual, language, and spatial cues. The predicate explains the relationship between the subject and object and can be categorized into different categories such as prepositions and verbs. A large visual gap exists although the visual relationship is included in the same predicate. This study improves upon a previous study (that uses language cues using two losses) and a spatial cue (that only includes individual information) by adding relative information on the subject and object of the extant study. The architectural limitation is demonstrated and is overcome to detect all zero-shot visual relationships. A new problem is discovered, and an explanation of how it decreases performance is provided. The experiment is conducted on the VRD and VG datasets and a significant improvement over previous results is obtained.

• Current Optics and Photonics
• /
• v.2 no.6
• /
• pp.547-553
• /
• 2018

In this paper, a new method for spatial filtering in digital holography is proposed and verified by simulations compared to conventional methods. The new method is based on the simultaneous acquisition of two digital holograms, which can be separated by distinct spatial modulation, in a single image. Two holograms are generated by two reference waves, which have different spatial modulation orientations. Then, the overlapping region between the DC term and the object wave in the first hologram can be replaced with a less-overlapping region of the object wave in the second hologram because the whole image contains two holograms where the same objective wave has been recorded. In the simulation results, it is confirmed that the reconstructed image by the new method has better quality than for the original method.

• Journal of Biomedical Engineering Research
• /
• v.28 no.1
• /
• pp.95-101
• /
• 2007

Recently, small-animal imaging technology has been rapidly developed for longitudinal screening of laboratory animals such as mice and rats. One of newly developed imaging modalities for small animals is an x-ray micro-CT (computed tomography). We have developed two types of x-ray micro-CT systems for small animal imaging. Both systems use flat-panel x-ray detectors and micro-focus x-ray sources to obtain high spatial resolution of $10{\mu}m$. In spite of the relatively large field-of-view (FOV) of flat-panel detectors, the spatial resolution in the whole-body imaging of rats should be sacrificed down to the order of $100{\mu}m$ due to the limited number of x-ray detector pixels. Though the spatial resolution of cone-beam CTs can be improved by moving an object toward an x-ray source, the FOV should be reduced and the object size is also limited. To overcome the limitation of the object size and resolution, we introduce zoom-in micro-tomography for high-resolution imaging of a local region-of-interest (ROI) inside a large object. For zoom-in imaging, we use two kinds of projection data in combination, one from a full FOV scan of the whole object and the other from a limited FOV scan of the ROI. Both of our micro-CT systems have zoom-in micro-tomography capability. One of both is a micro-CT system with a fixed gantry mounted with an x-ray source and a detector. An imaged object is laid on a rotating table between a source and a detector. The other micro-CT system has a rotating gantry with a fixed object table, which makes whole scans without rotating an object. In this paper, we report the results of in vivo small animal study using the developed micro-CTs.

• Proceedings of the KSRS Conference
• /
• 1999.11a
• /
• pp.194-197
• /
• 1999

In AVL, GIS analyze the information from the vehicles to provide commercial or other value far user. As spatial analysis functions in GIS make a new valuable information using the vehicle's position and geographic object's location, they perform an important roles to improve the management efficiency of vehicles. Most GIS however are used static data for the spatial analysis, so the research area on AVL used dynamic vehicle location has generated unsuitable result. In this study, we use GPS real time tracking data to perform spatial analysis between moving vehicle and static geographic object. The method proposed in this paper considers the driving direction of vehicle and creates the result which is located in forward of vehicle. In this paper, two spatial analysis functions, near and connectivity, are developed.

• Proceedings of the KSRS Conference
• /
• 2002.10a
• /
• pp.157-162
• /
• 2002

Recently, location based services, which make use of location information of moving objects, have obtained increasingly high attention. The moving objects are time-evolving spatial objects, that is, their locations are dynamically changed as time varies. Generally, GIS server stores and manages the spatial objects, of which locations are rarely changed. The traditional GIS server, however, has a difficulty to manage the moving objects, due to the fact of locations being frequently changed and the trajectory information (past locations of moving objects) being managed. In this paper, we have designed a moving object server, which stores and manages the locations in order to support various location based services. The moving object server is composed of a location acquisition component, a location storage component, and a location query component. The contribution of this paper is that we integrate the each work for location acquisition, storage, and query into a moving objects server.

• Journal of KIISE
• /
• v.42 no.3
• /
• pp.390-398
• /
• 2015

Spatial web objects refer to web documents that contain geographic information. Recently, services that create spatial web objects have increased greatly because of the advancements in devices such as smartphones. For services such as Twitter or Facebook, simple texts posted by users is stored along with information about the post's location. To search for such spatial web objects, a method that uses spatial information and text information simultaneously is required. Conventional spatial web object search methods mostly use R-tree and inverted file methods. However, these methods have a disadvantage of requiring a large volume of space when building indices. Furthermore, such methods are efficient for searching with many keywords but are inefficient for searching with a few keywords.. In this paper, we propose a spatial web object search method that uses a quad-tree and a patricia-trie. We show that the proposed technique is more effective than existing ones in searching with a small number of keywords. Furthermore, we show through an experiment that the space required by the proposed technique is much smaller than that required by existing ones.

• Journal of KIISE:Computing Practices and Letters
• /
• v.6 no.1
• /
• pp.1-10
• /
• 2000

Spatial analysis in GIS provides an important way that helps the end-users decision-making. For such a reason, query model for analysis should be able to support the users view conceptually in constructing query statements. The traditional approaches in design of query model used to extend the functionality of model that basically designed for manipulation of attribute-operations by appending operators for spatial operations to the query statements of model. However, by the reason of spatial operation's characteristics that are different from those of attribute operations In nature, the structures of query statements in previous approaches are unnatural, inconsistent, and therefore those query models in previous approaches are not able to support the users view in retrieving analysis. In this paper, we proposed the methodology for constructing of user query and internal processing this query based on object-oriented paradigm, in the view of spatial operations by using the basic concept that spatial query is a methodology for spatial analysis. In addition, we presented a strong possibility of designing spatial query model that might actively have interaction with its user by implementing CIW(Class-Information Window) query interface corresponded with the methodology proposed in this paper.

• Journal of KIISE:Computing Practices and Letters
• /
• v.8 no.6
• /
• pp.692-704
• /
• 2002

In this paper, we present am image retrieval method based on color-spatial features using quadtree and texture information extracted from object MBRs in an image. Tile proposed method consists of creating a DC image from an original image, changing a color coordinate system, and decomposing regions using quadtree. As such, conditions are present to decompose the DC image, then the system extracts representative colors from each region. And, image segmentation is used to search for object MBRs, including object themselves, object included in the background, or certain background region, then the wavelet coefficients are calculated to provide texture information. Experiments were conducted using the proposed similarity method based on color-spatial and texture features. Our method was able to refute the amount of feature vector storage by about 53%, but was similar to the original image as regards precision and recall. Furthermore, to make up for the deficiency in using only color-spatial features, texture information was added and the results showed images that included objects from the query images.

• Journal of Korea Multimedia Society
• /
• v.19 no.2
• /
• pp.112-121
• /
• 2016

ADAS(Advanced Driver Assistance Systems) and street furniture information collecting car like as MMS(Mobile Mapping System), they require object location estimation method for recognizing spatial information of object in road images. But, the case of conventional methods, these methods require additional hardware module for gathering spatial information of object and have high computational complexity. In this paper, for a coordinate of road sign in single camera image, a position estimation scheme of object in road images is proposed using the relationship between the pixel and object size in real world. In this scheme, coordinate value and direction are used to get coordinate value of a road sign in images after estimating the equation related on pixel and real size of road sign. By experiments with test video set, it is confirmed that proposed method has high accuracy for mapping estimated object coordinate into commercial map. Therefore, proposed method can be used for MMS in commercial region.

• Spatial Information Research
• /
• v.20 no.2
• /
• pp.165-172
• /
• 2012

The query of the K closest object pairs between two object sets frequently occurs at recently retrieval systems. The circular location property of objects should be considered for efficiently process queries finding such a K nearest object pair. In this paper, we propose the optimal algorithm finding the K object pairs which are closest to each other in a search space with a circular domain and show its performance by experiments. The proposed algorithm optimizes the cost of finding the K nearest object pairs by using the circular search distances which is much applied the circular location property.