• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.45 no.6
• /
• pp.170-178
• /
• 2008

A high performance object recognition algorithm using Fourier description of the 3D information of the objects is proposed. Object boundaries contain sufficient information for recognition in most of objects. However, it is not well utilized as the key solution of the object recognition since obtaining the accurate boundary information is not easy. Also, object boundaries vary highly depending on the size or orientation of object. The proposed object recognition algorithm is based on 1) the accurate object boundaries extracted from the 3D shape which is obtained by the laser scan device, and 2) reduction of the required database using the size and rotational invariant feature of the Fourier Descriptor. Such Fourier information is compared with the database and the recognition is done by selecting the best matching object. The experiments have been done on the rich database of MPEG 7 Part B.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2002.11a
• /
• pp.147-153
• /
• 2002

Due to the growth of population and industrial development at the coastal cities, there has been much increase in necessity to effective control of the wastes into the coastal water and river. The amount of disposal at those waters has been increased rapidly and it is necessary for us to track of it in order to keep the water clean. The investigation and research related to the water quality in this region have been conducted continuously but the systematic survey of the disposed wastes at the bottom was neglected and/or minor. In this study we surveyed the status of disposed waste distribution at the bottom coastal water and river from the scanned images. The intensity of sound received by the side scan sonar tow vehicle from the sea floor provides information as to the general distribution and characteristics of the superficial wastes. The port and starboard side scanned images produced from a transducer borne on a tow fish connected by tow cable to a tug boat have the area with width of 22m∼112m, and band of 44m∼224m. All data are displayed in real-time on a high-resolution color display (1280 ${\times}$ 1024 pixels) together with position information by DGPS. From the field measurement and analysis of the recorded images, we could draw the location and distribution of bottom disposals. Furthermore, we made a database system which might be fundamental for planning the waste reception and process control system.

• The KIPS Transactions:PartB
• /
• v.14B no.4
• /
• pp.249-254
• /
• 2007

A mobile robot with various types of sensors and wireless camera is introduced. We show this mobile robot can detect objects well by combining the results of active sensors and image processing algorithm. First, to detect objects, active sensors such as infrared rays sensors and supersonic waves sensors are employed together and calculates the distance in real time between the object and the robot using sensor's output. The difference between the measured value and calculated value is less than 5%. We focus on how to detect a object region well using image processing algorithm because it gives robots the ability of working for human. This paper suggests effective visual detecting system for moving objects with specified color and motion information. The proposed method includes the object extraction and definition process which uses color transformation and AWUPC computation to decide the existence of moving object. Shape information and signature algorithm are used to segment the objects from background regardless of shape changes. We add weighing values to each results from sensors and the camera. Final results are combined to only one value which represents the probability of an object in the limited distance. Sensor fusion technique improves the detection rate at least 7% higher than the technique using individual sensor.

• Journal of KIISE:Databases
• /
• v.35 no.6
• /
• pp.481-494
• /
• 2008

In this paper we propose Fourier magnitudes based privacy preserving clustering on time-series data. The previous privacy-preserving method, called DFT coefficient method, has a critical problem in privacy-preservation itself since the original time-series data may be reconstructed from privacy-preserved data. In contrast, the proposed DFT magnitude method has an excellent characteristic that reconstructing the original data is almost impossible since it uses only DFT magnitudes except DFT phases. In this paper, we first explain why the reconstruction is easy in the DFT coefficient method, and why it is difficult in the DFT magnitude method. We then propose a notion of distance-order preservation which can be used both in estimating clustering accuracy and in selecting DFT magnitudes. Degree of distance-order preservation means how many time-series preserve their relative distance orders before and after privacy-preserving. Using this degree of distance-order preservation we present greedy strategies for selecting magnitudes in the DFT magnitude method. That is, those greedy strategies select DFT magnitudes to maximize the degree of distance-order preservation, and eventually we can achieve the relatively high clustering accuracy in the DFT magnitude method. Finally, we empirically show that the degree of distance-order preservation is an excellent measure that well reflects the clustering accuracy. In addition, experimental results show that our greedy strategies of the DFT magnitude method are comparable with the DFT coefficient method in the clustering accuracy. These results indicate that, compared with the DFT coefficient method, our DFT magnitude method provides the excellent degree of privacy-preservation as well as the comparable clustering accuracy.

• KIPS Transactions on Software and Data Engineering
• /
• v.6 no.4
• /
• pp.203-210
• /
• 2017

In this paper, we propose an effective neural network model for image caption generation and model transfer. This model is a kind of multi-modal recurrent neural network models. It consists of five distinct layers: a convolution neural network layer for extracting visual information from images, an embedding layer for converting each word into a low dimensional feature, a recurrent neural network layer for learning caption sentence structure, and a multi-modal layer for combining visual and language information. In this model, the recurrent neural network layer is constructed by LSTM units, which are well known to be effective for learning and transferring sequence patterns. Moreover, this model has a unique structure in which the output of the convolution neural network layer is linked not only to the input of the initial state of the recurrent neural network layer but also to the input of the multimodal layer, in order to make use of visual information extracted from the image at each recurrent step for generating the corresponding textual caption. Through various comparative experiments using open data sets such as Flickr8k, Flickr30k, and MSCOCO, we demonstrated the proposed multimodal recurrent neural network model has high performance in terms of caption accuracy and model transfer effect.

• Journal of KIISE:Software and Applications
• /
• v.32 no.3
• /
• pp.202-214
• /
• 2005

The semantic web is the web paradigm that represents not general link of documents but semantics and relation of document. In addition it enables software agents to understand semantics of documents. We propose a semantic search based on inference with ontologies, which has the following characteristics. First, our search engine enables retrieval using explicit ontologies to reason though a search keyword is different from that of documents. Second, although the concept of two ontologies does not match exactly, can be found out similar results from a rule based translator and ontological reasoning. Third, our approach enables search engine to increase accuracy and precision by using explicit ontologies to reason about meanings of documents rather than guessing meanings of documents just by keyword. Fourth, domain ontology enables users to use more detailed queries based on ontology-based automated query generator that has search area and accuracy similar to NLP. Fifth, it enables agents to do automated search not only documents with keyword but also user-preferable information and knowledge from ontologies. It can perform search more accurately than current retrieval systems which use query to databases or keyword matching. We demonstrate our system, which use ontologies and inference based on explicit ontologies, can perform better than keyword matching approach .

• Economic and Environmental Geology
• /
• v.42 no.3
• /
• pp.273-282
• /
• 2009

To render more valuable information, a spatial database is being constructed from digitalized maps in the geographic areas. Transferring file-based maps into a spatial database, facilitates the integration of larger databases and information retrieval using database functions. Geological mapping is the graphical interpretation results of the geological phenomenon by geological surveyors, which is different from other thematic maps produced quantitatively. These features make it difficult to construct geologic databases needing geologic interpretation about various meanings. For those reasons, several organizations in the USA and Australia are suggesting the data model for the database construction. But, it is hard to adapt to a domestic environment because of the representation differences of geological phenomenon. This paper suggests the data model adaptive in domestic environment analyzing 1:50,000 scales of geologic maps and more detailed mine geologic maps. The suggested model is a logical data model for the ArcGIS GeoDatabase. Using the model it can be efficiently applicable in the 1:50,000 scales of geological maps. It is expected that the geologic data model suggested in this paper can be used for integrated use and efficient management of geologic maps.

• Cartoon and Animation Studies
• /
• s.45
• /
• pp.27-50
• /
• 2016

The present study argues that documentary-animation films, which are based on actual human voices, on the level of representation, constitute a new expansion for the medium of animation films, which serve as testimonies to the real world. Animation films are produced using very diverse techniques so that they are complex to the degree of being indefinable, and documentary films, though based on objective representation, increase in complexity in that there exist various types of artificial interventions such as direction and digital image processing. Having emerged as a hybrid genre of the two media, documentary-animation films draw into themselves actual events and elements so that they conceptually share reality-based narratives and are visually characterized by the trappings of animation films. Generally classified as 'animated documentaries', this genre triggered discussions following the release of , a work that is mistaken as having used rotoscoping transforming live action in terms of the technique. When analyzed in detail, however, this work is presented as an ambiguous medium where the characteristics of animation films, which are virtual simulacra without reality, and of documentaries, which are based on the objective indexicality of the referents, coexist because of its mixed use of typical animation techniques, 3D programs, and live-action images. Discussed in the present study, , , and share the characteristics of the medium of documentaries in that the narratives develop as testimonies of historical figures but, at the same time, are connected to animation films because of their production techniques and direction characteristics. Consequently, this medium must be discussed as a new expansion rather than being included in the existing classification system, and such a presupposition is an indispensable process for directly facing the reality of the works and for developing discussions. Through works that directly use the interviewees' voices yet do not transcend the characteristics of animation films, the present study seeks to define documentary-animation films and to discuss the possibility of the medium, which has expanded as a testimony to the real world.

• Ecology and Resilient Infrastructure
• /
• v.3 no.4
• /
• pp.307-314
• /
• 2016

River restoration has recently been performed not only for the improvement of the artificial parts in the past but also for the restoration of abandoned river reaches which were blocked and isolated. For the restoration of abandoned river reaches, it is important to recover the hydraulic and ecological connectivity in the isolated space by longitudinal structures like levees. But because the assessment tools to determine whether the river restoration is performed properly are so rare at present, we aim to provide a tool for assessing ecological connectivity in a target river in this study. In the first step, one-dimensional numerical model for rainfall-runoff and channel routing was developed and then applied to the watershed of the Cheongmi Stream. In this step, a numerical model was developed to assess the restoration of connectivity. The model consists of two parts: one part is to convert the results of one-dimensional channel routing into two-dimensional spatial distribution. The other is to calculate the habitat suitability index according to time steps by using two-dimensional hydraulic features. The model was applied to a restoration area of the Cheongmi Stream. The advantage of this study is that two-dimensional hydraulic analysis can be easily obtained from one-dimensional hydraulic analysis without a complex and time-consuming two-dimensional analysis. HHS (Hydraulic Habitat Suitablility) by sections of target reaches and target species can be easily obtained using the results of this study.

• Korean Journal of Remote Sensing
• /
• v.35 no.6_1
• /
• pp.959-971
• /
• 2019

CNN (Convolutional Neural Network) is one representative deep learning algorithm, which can extract high-level spatial and spectral features, and has been applied for hyperspectral image classification. However, one significant drawback behind the application of CNNs in hyperspectral images is the high dimensionality of the data, which increases the training time and processing complexity. To address this problem, several CNN based hyperspectral image classification studies have exploited PCA (Principal Component Analysis) for dimensionality reduction. One limitation to this is that the spectral information of the original image can be lost through PCA. Although it is clear that the use of PCA affects the accuracy and the CNN training time, the impact of PCA for CNN based hyperspectral image classification has been understudied. The purpose of this study is to analyze the quantitative effect of PCA in CNN for hyperspectral image classification. The hyperspectral images were first transformed through PCA and applied into the CNN model by varying the size of the reduced dimensionality. In addition, 2D-CNN and 3D-CNN frameworks were applied to analyze the sensitivity of the PCA with respect to the convolution kernel in the model. Experimental results were evaluated based on classification accuracy, learning time, variance ratio, and training process. The size of the reduced dimensionality was the most efficient when the explained variance ratio recorded 99.7%~99.8%. Since the 3D kernel had higher classification accuracy in the original-CNN than the PCA-CNN in comparison to the 2D-CNN, the results revealed that the dimensionality reduction was relatively less effective in 3D kernel.