Virtual crop models make great significances to the research of quantification of field evapotranspiration, design of plant type, and optimization of cultural practices. Because most existing plant-modeling systems require the user to have specific biological knowledge about plants, thus it is often a hard and laborious task by using these systems to create a three-dimensional plant model. Moreover, there is lack of software tools to specially simulate the structures and growth process of a certain kind of crop. With regard to those statuses of virtual plant modeling, in this article a powerful and convenient tomato plant modeling system is developed with the integrated platform of Visual C++ 6.0 and OpenGL. Based on the growth laws and the topological structure's character of tomato plant, Parametric Lsystem is used to simulate the topological structures of tomato plant. The technique of parametric triangular meshes is used to model different organs of tomato, and the surface-rendering techniques in Computer Graphics are used to render the created shapes of tomato plant at both organ and individual levels. Based on the accumulated temperature model, the system can predict the growth status of tomato plant, and give some suggestion of tomato planting.
How to select and combine good traits of rice to get high-production individuals is one of the key points in developing crop ideotype cultivation technologies. Existing cultivation methods for producing ideal plants, such as field trials and crop modeling, have some limits. In this paper, we propose a method based on a genetic algorithm (GA) and a functional-structural plant model (FSPM) to optimize plant types of virtual rice by dynamically adjusting phenotypical traits. In this algorithm, phenotypical traits such as leaf angles, plant heights, the maximum number of tiller, and the angle of tiller are considered as input parameters of our virtual rice model. We evaluate the photosynthetic output as a function of these parameters, and optimized them using a GA. This method has been implemented on GroIMP using the modeling language XL (eXtended L-System) and RGG (Relational Growth Grammar). A double haploid population of rice is adopted as test material in a case study. Our experimental results show that our method can not only optimize the parameters of rice plant type and increase the amount of light absorption, but can also significantly increase crop yield.
Previously, convolutional neural networks mostly used deep semantic feature information obtained from several convolutions for image classification. Such deep semantic features have a larger receptive field, and the features extracted are more effective as the number of convolutions increases, which helps in the classification of targets. However, this method tends to lose the shallow local features, such as the spatial connectivity and correlation of tumor region texture and edge contours in breast histopathology images, which leads to its recognition accuracy not being high enough. To address this problem, we propose a multi-level feature fusion method for breast histopathology image classification. First, we fuse shallow features and deep semantic features by attention mechanism and convolutions. Then, a new weighted cross entropy loss function is used to deal with the misjudgment of false negative and false positive. And finally, the correlation of spatial information is used to correct the misjudgment of some patches. We have conducted experiments on our own datasets and compared with the base network Inception-ResNet-v2, which has a high accuracy. The proposed method achieves an accuracy of 99.0% and an AUC of 99.9%.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.