The problem of domestic refuse is becoming more and more serious with the use of all kinds of equipment in medical institutions.This matter arouses people’s attention.Traditional artificial waste classification is su...The problem of domestic refuse is becoming more and more serious with the use of all kinds of equipment in medical institutions.This matter arouses people’s attention.Traditional artificial waste classification is subjective and cannot be put accurately;moreover,the working environment of sorting is poor and the efficiency is low.Therefore,automated and effective sorting is needed.In view of the current development of deep learning,it can provide a good auxiliary role for classification and realize automatic classification.In this paper,the ResNet-50 convolutional neural network based on the transfer learning method is applied to design the image classifier to obtain the domestic refuse classification with high accuracy.By comparing the method designed in this paper with back propagation neural network and convolutional neural network,it is concluded that the CNN based on transfer learning method applied in this paper with higher accuracy rate and lower false detection rate.Further,under the shortage situation of data samples,the method with transfer learning and ResNet-50 training model is effective to improve the accuracy of image classification.展开更多
The sustainability of ecosystem restoration of refuse dumps in open-pit coal mines depends on plant species selection, their configuration, and the optimal usage of water resources. This study is based on field experi...The sustainability of ecosystem restoration of refuse dumps in open-pit coal mines depends on plant species selection, their configuration, and the optimal usage of water resources. This study is based on field experiments in the northern refuse dump of the Heidaigou open-pit coal mine in Inner Mongolia of China established in 1995. Eight plant configurations, including trees, shrubs, grasses, and their combinations, as well as the adjacent community of natural vegetation, were selected. The succession of the revegetated plants, soil water storage, the spatiotemporal distribution of plant water deficits degree and its compensation degree were also studied. Results indicated that the vegetation cover （shrubs and herbaceous cover）, richness, abundance, soil nutrients （soil organic matter, N and P）, and biological soil crust coverage on the soil surface are significantly influenced by the vegetation configurations. The average soil water storage values in the shrub ＋ grass and grass communities throughout the growing season are 208.69 mm and 206.55 mm, which are the closest to that of in the natural vegetation community （215.87 mm）. Plant water deficits degree in the grass and shrub ＋ grass communities were the lowest, but the degrees of water deficit compensation in these configuration were larger than those of the other vegetation configurations. Differences in plant water deficit degree and water compensation among the different config- urations were significant （P 〈0.05）. Plant water deficit degrees were predominantly minimal on the surface, increased with increasing soil depth, and remained stable at 80 cm soil depth. The soil moisture compensation in the natural vegetation, shrub ＋ grass, and grass communities changed at 10%, while that in other vegetation communities changed between 20% and 40%. Overall, we conclude that the shrub ＋ grass and grass configuration modes are the optimal vegetation restoration models in terms of ecohydrology for future ecological engineering projects.展开更多
This study focused on two woody leafy vegetables Leptadenia hastata Decne and Senna obtusifolia Link, commonly consumed in Senegal. Leaves were col-lected from three regions. Then, proximate analyses and micronutrient...This study focused on two woody leafy vegetables Leptadenia hastata Decne and Senna obtusifolia Link, commonly consumed in Senegal. Leaves were col-lected from three regions. Then, proximate analyses and micronutrients were carried out to evaluate their nutritional values. Results revealed that protein level of S. obtusifolia (SO) is richer (21.75%) than Leptadenia hastata (LH) (18.16%). The cellulose and carbohydrate contents of the two vegetable’s leaves are in the same order except those of LH from Widou which are less rich in cellulose (8.31%) and richest in carbohydrate (6.35%). These leaves are also good sources of various mineral elements and especially iron. Leaves of LH appear to be richer in iron and magnesium, while SO appears to be richer in calcium. Vitamin C intakes of SO leaves are better than those of LH and respectively range from 142 to 196.5 and 22.5 to 159.5 mg/100 g. According to the use of this leafy vegetable by the populations, a domestication opportunity is thus justified to ensure availability and accessibility of these significant sources of micronutrients.展开更多
基金This work was supported in part by the National Natural Science Foundation of China under Grant 61806028,Grant 61672437 and Grant 61702428Sichuan Science and Technology Program under Grants 21ZDYF2484,2021YFN0104,21GJHZ0061,21ZDYF3629,21ZDYF2907,21ZDYF0418,21YYJC1827,21ZDYF3537,2019YJ0356the Chinese Scholarship Council under Grants 202008510036,201908515022.
文摘The problem of domestic refuse is becoming more and more serious with the use of all kinds of equipment in medical institutions.This matter arouses people’s attention.Traditional artificial waste classification is subjective and cannot be put accurately;moreover,the working environment of sorting is poor and the efficiency is low.Therefore,automated and effective sorting is needed.In view of the current development of deep learning,it can provide a good auxiliary role for classification and realize automatic classification.In this paper,the ResNet-50 convolutional neural network based on the transfer learning method is applied to design the image classifier to obtain the domestic refuse classification with high accuracy.By comparing the method designed in this paper with back propagation neural network and convolutional neural network,it is concluded that the CNN based on transfer learning method applied in this paper with higher accuracy rate and lower false detection rate.Further,under the shortage situation of data samples,the method with transfer learning and ResNet-50 training model is effective to improve the accuracy of image classification.
基金supported by the CAS Action-plan for Western Development(KZCX2-XB3-13-03)Chinese National Natural Scientific Foundation(4120108431170385)
文摘The sustainability of ecosystem restoration of refuse dumps in open-pit coal mines depends on plant species selection, their configuration, and the optimal usage of water resources. This study is based on field experiments in the northern refuse dump of the Heidaigou open-pit coal mine in Inner Mongolia of China established in 1995. Eight plant configurations, including trees, shrubs, grasses, and their combinations, as well as the adjacent community of natural vegetation, were selected. The succession of the revegetated plants, soil water storage, the spatiotemporal distribution of plant water deficits degree and its compensation degree were also studied. Results indicated that the vegetation cover （shrubs and herbaceous cover）, richness, abundance, soil nutrients （soil organic matter, N and P）, and biological soil crust coverage on the soil surface are significantly influenced by the vegetation configurations. The average soil water storage values in the shrub ＋ grass and grass communities throughout the growing season are 208.69 mm and 206.55 mm, which are the closest to that of in the natural vegetation community （215.87 mm）. Plant water deficits degree in the grass and shrub ＋ grass communities were the lowest, but the degrees of water deficit compensation in these configuration were larger than those of the other vegetation configurations. Differences in plant water deficit degree and water compensation among the different config- urations were significant （P 〈0.05）. Plant water deficit degrees were predominantly minimal on the surface, increased with increasing soil depth, and remained stable at 80 cm soil depth. The soil moisture compensation in the natural vegetation, shrub ＋ grass, and grass communities changed at 10%, while that in other vegetation communities changed between 20% and 40%. Overall, we conclude that the shrub ＋ grass and grass configuration modes are the optimal vegetation restoration models in terms of ecohydrology for future ecological engineering projects.
文摘This study focused on two woody leafy vegetables Leptadenia hastata Decne and Senna obtusifolia Link, commonly consumed in Senegal. Leaves were col-lected from three regions. Then, proximate analyses and micronutrients were carried out to evaluate their nutritional values. Results revealed that protein level of S. obtusifolia (SO) is richer (21.75%) than Leptadenia hastata (LH) (18.16%). The cellulose and carbohydrate contents of the two vegetable’s leaves are in the same order except those of LH from Widou which are less rich in cellulose (8.31%) and richest in carbohydrate (6.35%). These leaves are also good sources of various mineral elements and especially iron. Leaves of LH appear to be richer in iron and magnesium, while SO appears to be richer in calcium. Vitamin C intakes of SO leaves are better than those of LH and respectively range from 142 to 196.5 and 22.5 to 159.5 mg/100 g. According to the use of this leafy vegetable by the populations, a domestication opportunity is thus justified to ensure availability and accessibility of these significant sources of micronutrients.
