Jun-Soo Kim

Learning-based approach for license plate recognition

Presents a learning-based approach for the construction of a license-plate recognition system. The system consists of three modules. They are, respectively, the car detection module, the license-plate segmentation module and the recognition module. The car detection module detects a car in a given image sequence obtained from a camera with a simple color-based approach. The segmentation module extracts the license plate in the detected car image using neural networks as filters for analyzing the color and texture properties of the license plate. The recognition module then reads the characters on the detected license plate with a support vector machine (SVM)-based character recognizer. The system has been tested with 1000 video sequences obtained from toll-gates, parking lots, etc., and has shown the following performances on average: car detection rate 100%, segmentation rate 97.5%, and character recognition rate about 97.2%.

Initial Responses of Quercus serrata Seedlings and Forest Understory to Experimental Gap Treatments

Pinus thunbergii plantations in Pohang-si, Gyeongsangbuk-do, Korea, are of low ecological quality, with arrested succession and a high proportion of ruderal species. To improve the quality of the habitat, we created canopy gaps () and monitored changes in abiotic (light availability, canopy openness) and biotic (survival and growth of seedlings and understory communities) variables in 2007 and 2008 in plots that had received one of five types of treatment: cutting of canopy trees and removal of the understory (CU), cutting of canopy trees only (C), girdling of canopy trees and removal of the understory (GU), girdling of canopy trees (G) or control. Each treatment was applied to three replicate plots. Abiotic variables did not significantly differ among treatments. Survival rates of target species were slightly lower in the CU, G and control conditions. Based on logistic regression analysis, the only significant growth factor affecting survival was height growth. Positive effects of seedling height and leaf area growth on survival were also detected, but did not reach statistical significance. In treatment G, gradual improvement of overstory conditions and mitigation of competition by limitation of disturbance to the understory community were likely to have promoted seedling growth. There were no significant effects of gap treatments on changes in species abundance (cover and richness) and composition of understory between the study years. This result implies that the small gaps created in our study may be below the threshold size to affect understory growth. However, the results of this study are based on a short-term investigation of only two years. Long-term research is strongly recommended to clarify the effects of gap treatment on plant communities in afforested areas.