• The Journal of the Korea Contents Association
• /
• v.12 no.11
• /
• pp.1-8
• /
• 2012

The application of object recognition technology has been increased with a growing need to introduce automated system in industry. However, object transformed by noises and shadows appeared from illumination causes challenge problem in object detection and recognition. In this paper, an illumination invariant object detection using a DoG filter and adaptive threshold is proposed that reduces noises and shadows effects and reserves geometry features of object. And also, rotation invariant object recognition is proposed that has trained with neural network using classes categorized by object type and rotation angle. The simulation has been processed to evaluate feasibility of the proposed method that shows the accuracy of 99.86% and the matching speed of 0.03 seconds on ETRI database, which has 16,848 object images that has obtained in various lighting environment.

• The KIPS Transactions:PartC
• /
• v.13C no.3 s.106
• /
• pp.383-388
• /
• 2006

Most of GIS services have been operated in single applications, and as data processing and computer and mobile technologies have developed rapidly, users request for efficient sharing between each GIS's own data and various different GIS's. However, since many GIS applications maintain their own data formats, they are incapable of processing data formats different with each other, and do not have a filtering function for mobile GIS. This paper designs an integrated preprocessor, SMP to accept features of various current formats of geographic information such as DXF(Drawing eXchange Format), DWG(DraWinG), SHP(SHaPefile), etc., and to extract core information for describing maps. The geographic information extracted by SMP(Synthetic Map Preprocessor) shows consistency in various formats by visualizing through the integrated view. By generating the extracted core data in GML, it supports rapid access to mobile devices and extensibility of file formats overcoming heterogeneity.

• Korean Journal of Ichthyology
• /
• v.20 no.4
• /
• pp.272-278
• /
• 2008

Optimum concentration of alizarine complexone (ALC) for staining of otoliths for Pacific cod was determined by examination of stained quality, growth and survival rates. Small juveniles ($34.1{\pm}1.8mm$ TL, $mean{\pm}SD$) were immersed in four different concentrations (20, 40, 60, 80 mg/L) and large ones ($73.8{\pm}1.2mm$ TL, $mean{\pm}SD$) in five different concentrations (1, 5, 10, 20, 40 mg/L) of ALC for 24 hrs. The fish were then reared and observed for 14 days. The stained quality of otoliths was evaluated by dividing them into four grades by the presence of fluorescent stains on the otoliths under G-excited UV light. The small juveniles stained with the different ALC concentrations had 100% staining success, and their growth rate and body weight did not differ significantly between treatments. Optimal ALC concentration for the smaller juveniles was 20 mg/L in the present experiment. On the other hand, distinct fluorescent rings were confirmed on otoliths of larger juveniles in the 10 mg/L treatment with high survival rate. These results suggest that ALC staining at concentrations of 10 mg/L with larger juveniles will be helpful for release-recapture experiments for stock enhancement of Pacific cod.