Especially in recent years, deep learning has become a very effective tool for object identification. However, in general, the automatic object identification tends not to work well on ambiguous, amorphous objects suc...Especially in recent years, deep learning has become a very effective tool for object identification. However, in general, the automatic object identification tends not to work well on ambiguous, amorphous objects such as vegetation. In this study, we developed a simple but effective approach to identify ambiguous objects and applied the method to several moss species. The technique called chopped picture method, where teacher images are systematically dissected into numerous small squares. As a result, the model correctly classified 3 moss species and “non-moss” objects in test images with accuracy more than 90%. Using this approach will help progress in computer vision studies for various ambiguous objects.展开更多
Biogeochemical feedback processes between soil organic carbon (SOC) in high-latitude organic soils and climate change is of great concern for projecting future climate. More accurate models of the SOC stock and its dy...Biogeochemical feedback processes between soil organic carbon (SOC) in high-latitude organic soils and climate change is of great concern for projecting future climate. More accurate models of the SOC stock and its dynamics in organic soil are of increasing importance. As a first step toward creating a soil model that accurately represents SOC dynamics, we have created the Physical and Biogeochemical Soil Dynamics Model (PB-SDM) that couples a land surface model with a SOC dynamics model to simulate the feedback cycle of SOC accumulation and thermal hydrological dynamics of high-latitude soils. The model successfully simulated soil temperatures for observed data from a boreal forest near Fairbanks, and 2000 year simulations indicated that the effect of the feedback cycle of SOC accumulation on soil thickness would result in a significant differences in the amount of SOC.展开更多
文摘Especially in recent years, deep learning has become a very effective tool for object identification. However, in general, the automatic object identification tends not to work well on ambiguous, amorphous objects such as vegetation. In this study, we developed a simple but effective approach to identify ambiguous objects and applied the method to several moss species. The technique called chopped picture method, where teacher images are systematically dissected into numerous small squares. As a result, the model correctly classified 3 moss species and “non-moss” objects in test images with accuracy more than 90%. Using this approach will help progress in computer vision studies for various ambiguous objects.
文摘Biogeochemical feedback processes between soil organic carbon (SOC) in high-latitude organic soils and climate change is of great concern for projecting future climate. More accurate models of the SOC stock and its dynamics in organic soil are of increasing importance. As a first step toward creating a soil model that accurately represents SOC dynamics, we have created the Physical and Biogeochemical Soil Dynamics Model (PB-SDM) that couples a land surface model with a SOC dynamics model to simulate the feedback cycle of SOC accumulation and thermal hydrological dynamics of high-latitude soils. The model successfully simulated soil temperatures for observed data from a boreal forest near Fairbanks, and 2000 year simulations indicated that the effect of the feedback cycle of SOC accumulation on soil thickness would result in a significant differences in the amount of SOC.