• Journal of the Korean Society of Marine Environment & Safety
• /
• v.30 no.2
• /
• pp.147-156
• /
• 2024

The sustainable management and enhancement of marine resources are becoming increasingly important issues worldwide. This study was conducted in response to these challenges, focusing on the development and performance comparison of fish detection and classification models as part of a deep learning-based technique for assessing the effectiveness of marine resource enhancement projects initiated by the Korea Fisheries Resources Agency. The aim was to select the optimal model by training various sizes of YOLOv8-Seg models on a fish image dataset and comparing each performance metric. The dataset used for model construction consisted of 36,749 images and label files of 12 different species of fish, with data diversity enhanced through the application of augmentation techniques during training. When training and validating five different YOLOv8-Seg models under identical conditions, the medium-sized YOLOv8m-Seg model showed high learning efficiency and excellent detection and classification performance, with the shortest training time of 13 h and 12 min, an of 0.933, and an inference speed of 9.6 ms. Considering the balance between each performance metric, this was deemed the most efficient model for meeting real-time processing requirements. The use of such real-time fish detection and classification models could enable effective surveys of marine resource enhancement projects, suggesting the need for ongoing performance improvements and further research.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.6
• /
• pp.1237-1242
• /
• 2017

Computer vision is coming more interesting with new imaging sensors' new capabilities which enable it to understand more its surrounding environment by imitating human vision system with artificial intelligence techniques. In this paper, we made experiments with Kinect camera, a new depth sensor for object detection and distance measurement functions, most essential functions in computer vision such as for unmanned or manned vehicles, robots, drones, etc. Therefore, Kinect camera is used here to estimate the position or the location of objects in its field of view and measure the distance from them to its depth sensor in an accuracy way by checking whether that the detected object is real object or not to reduce processing time ignoring pixels which are not part of real object. Tests showed promising results with such low-cost range sensor, Kinect camera which can be used for object detection and distance measurement which are fundamental functions in computer vision applications for further processing.

• Journal of Digital Contents Society
• /
• v.18 no.6
• /
• pp.1017-1024
• /
• 2017

The development of technology has brought changes in content media. Starting from voice and sound media in the oral era, through text and painting, the realism has led to the development of visual media plus sound and image media. What we should consider here is not only the one-sided influence of change in the media due to the development of technology, but also the understanding, concentration, and commitment of information depending on which generation has access to the media Therefore, we focus on the digital native generation that uses digital as main media. The features of the digital native generation include the ability to process visual information quickly, multi-tasking, and divisionism. In this paper, we propose a moving pictogram for the digital native generation, and a moving pictogram for exit pictogram which shows limitation. The new dynamic pictograms that fit to the characteristics of the digital native generation, as well as interactive dynamic pictograms, are areas of thought and research on which this paper can be regarded as the first step.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.12 no.4
• /
• pp.306-310
• /
• 2002

In this paper, we realized a system that converts the Korean alphabet (Hangul) images, which are in any domain that is formed by the finger movement on the Hangul document, to the editable characters and then outputs them to the word editor. The domain of hand is separated from the sphere of document in the pre-process step of image. The centroid point of hand is drawn by the maximum circular movement method. After the system recognizes the hand with the circular pattern vector algorithm, finds out the position of finger by the distance spectrum and then draws out the sphere of selected character image by the finger movement to divide the characters into character units by applying the histogram between the Hangul characters. We standardized the characters of various sizes. We used the circular pattern vector algorithm that grafts on the fuzzy inference to divert the character images of the domain, which user wants, to the editable characters by comparing the characteristic vectors between the standard pattern character and the inputted character and by recognizing the character.

• Journal of KIISE:Computer Systems and Theory
• /
• v.35 no.4
• /
• pp.150-157
• /
• 2008

This paper presents a design and implementation of realtime video player that operates on a tiled-display system consisting of multiple PCs to provide a very large and high resolution display. In the proposed system, the master process transmits a compressed video stream to multiple PCs using UDP multicast. All slaves(PC) receive the same video stream, decompress, clip their designated areas from the decompressed video frame, and display it to their displays while being synchronized with each other. A simple synchronization mechanism based on the H/W clock of each slave is proposed to avoid the skew between the tiles of the display, and a flow-control mechanism based on the bit-rate of the video stream and a pre-buffering scheme are proposed to prevent the jitter The proposed system is implemented with Microsoft DirectX filter technology in order to decouple the video/audio codec from the player.

• The Journal of the Korea Contents Association
• /
• v.10 no.10
• /
• pp.172-179
• /
• 2010

The study on works motivated from interest in the nature of matters and inherent visual-perceptual structure in them aims at expressing formative continuity of the connections of three dimensions of simple geometrical shapes such as circles and lines, which are characteristics of shape of molecules. With such a purpose, this study examined the geometrical shapes in modern arts and structural connection and symbolism of molecule structure, and based on such considerations, it expressed successive formative beauty which comes from repetitive connection between units by creating stereogram of simple geometrical shapes of molecule structure. The types of works include a method of connecting the units of molecule models and molecules seen in electron microscope with lines as a parameter and connecting units directly, which are used to express body accessory and metallic sculptures. Consequently, it attempted formation occurring spatial composition of continuity of division and duplication through direct connection between units and circular continuity coming from connection of simple geometrical shapes of molecule images such as spheres and curves transformed into stereogram.

• Journal of KIISE:Computer Systems and Theory
• /
• v.36 no.3
• /
• pp.199-209
• /
• 2009

In this paper, we suggest a stroke level of detail (LOD) based on a saliency density. On painter]y rendering, the stroke LOD has an advantage of making the observer concentrate on the main object and improving accuracy of expression. For the stroke LOD, it is necessary to classify the detailed and abstracted area. We divide the area on the basis of saliency distribution and the level of detailed expression is controlled based on the saliency information. 'We define that the area of which the saliency distribution is high is a major subject that an artist tries to express, it is described in detail. The area of which the saliency distribution is low is abstractly described. Each divided area has the abstraction level. And by adapting the brushes of which sizes are appropriate to each level, it is possible to express the area which needs to be expressed in details from the one which needs to be expressed abstractly.

• Journal of Korea Multimedia Society
• /
• v.16 no.12
• /
• pp.1385-1392
• /
• 2013

There is a variety of researches about recognition system using biometric data these days. In this study, we propose a new algorithm, uses simultaneous equation that made of the edge of objects, to segment an iris region without threshold values from an anterior eye image. The algorithm attempts to find a center area through calculated outskirts information of an iris, and decides the area where the most points are accumulated. To verify the proposed algorithm, we conducted comparative experiments to Hough transform and Daugman's method, based on 50 images anterior eye images. It was found that proposed algorithm is 5 and 75 times faster than on each algorithm, and showed high accuracy of detecting a center point (95.36%) more than Hough transform (92.43%). In foreseeable future, this study is expected to useful application in diverse department of human's life, such as, identification system using an iris, diagnosis a disease using an anterior image.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.5 no.2
• /
• pp.54-68
• /
• 2002

In this paper, We propose a multi-dimensional index structure, called a VA (vector approximate) -tree that constructs a tree with vector approximates of multi-dimensional feature vectors. To save storage space for index structures, the VA-tree employs vector approximation concepts of VA-file that presents feature vectors with much smaller number of bits than original value. Since the VA-tree is a tree structure, it does not suffer from performance degradation owing to the increase of data. Also, even though the VA-tree is MBR Minimum Bounding Region) based tree structure like a R-tree, its split algorithm never allows overlap between MBRs. We show through various experiments that our proposed VA-tree is the efficient index structure for large amount of multi-dimensional data.

• Journal of Digital Contents Society
• /
• v.18 no.7
• /
• pp.1323-1331
• /
• 2017

Despite the fact that the copyright has grown into a large-scale business, there are many constant problems especially in image copyright. In this study, we propose an automatic object extraction and classification system for the scanned educational book image by combining document image processing and intelligent information technology like deep learning. First, the proposed technology removes noise component and then performs a visual attention assessment-based region separation. Then we carry out grouping operation based on extracted block areas and categorize each block as a picture or a character area. Finally, the caption area is extracted by searching around the classified picture area. As a result of the performance evaluation, it can be seen an average accuracy of 83% in the extraction of the image and caption area. For only image region detection, up-to 97% of accuracy is verified.