• Korean Journal of Environmental Agriculture
• /
• v.17 no.4
• /
• pp.358-361
• /
• 1998

This study was carried out to develop a short term hardening method of tree seedlings of Rhus chinensis Mill., Evodia daniellii Hemsley and Parthenocissus tricuspidata(Sieb. et Zucc.) Planck for afforestation on a concave and a crack of cut-rock slope. The seedlings were grown in a cylinder shaped pot made of polyvinyl net with the soil media of peatmoss, vermiculite, clay, compost, fertilizer, and absorbant(40:25:19:15:1:0.1, v:v). They were cultivated in a greenhouse for four months and in field condition for two months. During the last three months of the growing period the seedlings were hardened by periodic desiccation and irrigation in 4 to 10 days interval. The hardened seedlings showed lower leaf water potential, higher leaf osmotic pressure, and lower T/R ratio than those before the hardening. The hardened seedlings survived well on the soil medium in the concave of cut-rock slope.

• Journal of Korean Society of Forest Science
• /
• v.110 no.2
• /
• pp.155-164
• /
• 2021

Many studies have been conducted on developing automatic plant identification algorithms using machine learning to various plant features, such as leaves and flowers. Unlike other plant characteristics, barks show only little change regardless of the season and are maintained for a long period. Nevertheless, barks show a complex shape with a large variation depending on the environment, and there are insufficient materials that can be utilized to train algorithms. Here, in addition to the previously published bark image dataset, BarkNet v.1.0, images of barks were collected, and a dataset consisting of 53 tree species that can be easily observed in Korea was presented. A convolutional neural network (CNN) was trained and tested on the dataset, and the factors that interfere with the model's performance were identified. For CNN architecture, VGG-16 and 19 were utilized. As a result, VGG-16 achieved 90.41% and VGG-19 achieved 92.62% accuracy. When tested on new tree images that do not exist in the original dataset but belong to the same genus or family, it was confirmed that more than 80% of cases were successfully identified as the same genus or family. Meanwhile, it was found that the model tended to misclassify when there were distracting features in the image, including leaves, mosses, and knots. In these cases, we propose that random cropping and classification by majority votes are valid for improving possible errors in training and inferences.

• Journal of Korean Society of Transportation
• /
• v.20 no.1
• /
• pp.77-90
• /
• 2002

Network expansion has been an inevitable method for most traffic equilibrium assignments to consider intersection movements such as intersection delays. The drawback of network expansion is that because it dramatically increases network sizes to emulate possible directional movements as corresponding links, not only is complexities for building network amplified, but computational performance is shrunk. This paper Proposes a new variational inequality formulation for a user-optimal traffic equilibrium assignment model to explicitly consider directional delays without building expanded network structures. In the formulation, directional delay functions are directly embedded into the objective function, thus any modification of networks is not required. By applying a vine-based shortest Path algorithm into the diagonalization algorithm to solve the problem, it is additionally demonstrated that various loop-related movements such as U-Turn, P-Turn, etc., which are frequently witnessed near urban intersections, can also be imitated by blocking some turning movements of intersections. The proposed formulation expects to augment computational performance through reduction of network-building complexities.