In this study,a fast object detection algorithm based on binary deep convolution neural networks(CNNs)is proposed.Convolution kernels of different sizes are used to predict classes and bounding boxes of multi-scale ob...In this study,a fast object detection algorithm based on binary deep convolution neural networks(CNNs)is proposed.Convolution kernels of different sizes are used to predict classes and bounding boxes of multi-scale objects directly in the last feature map of a deep CNN.In this way,rapid object detection with acceptable precision loss is achieved.In addition,binary quantisation for weight values and input data of each layer is used to squeeze the networks for faster object detection.Compared to full-precision convolution,the proposed binary deep CNNs for object detection results in 62 times faster convolutional operations and 32 times memory saving in theory,what’s more,the proposed method is easy to be implemented in embedded computing systems because of the binary operation for convolution and low memory requirement.Experimental results on Pascal VOC2007 validate the effectiveness of the authors’proposed method.展开更多
In recent years,biochar has attracted considerable attention for soil quality improvement and carbon sequestration due to its unique physicochemical properties.However,the mechanism by which biochar application negati...In recent years,biochar has attracted considerable attention for soil quality improvement and carbon sequestration due to its unique physicochemical properties.However,the mechanism by which biochar application negatively affects the growth of crop seedlings has not been fully investigated.In this study,a hydroponic experiment was conducted to evaluate the response of rice,wheat,and corn seedlings to biochar application(CK,0 g/L;BC1,0.5 g/L;and BC2,1.0 g/L).Compared with the CK treatment,the BC1 and BC2 treatments decreased the fresh shoot and root weights of rice and corn seedlings(P<0.05),but there was no significant effect on wheat seedlings(P>0.05).For the contents of nutrient elements in seedlings,both BC1 and BC2 treatments hindered the roots from absorbing Fe and Cu and increased the uptake of Ca and Mn.Compared with the CK treatment,the translocation factor(TF)values of Ca,Mn,and Zn were significantly decreased especially in rice seedlings(35.3%-36.8%,68.7%-76.5%,and 29.8%-22.0%,respectively)under the BC1 and BC2 treatments,while only Mn was significantly decreased in wheat and corn seedlings(P<0.05).Transmission electron microscope(TEM)analysis of root cross-sections showed that nano-sized biochar particles(10~23 nm)were found in the root cells under BC2 treatment conditions.Our findings reveal that a large amount of biochar application can reduce nutrient absorption and translocation,and hinder rice,wheat,and corn seedlings,particularly rice seedling,in hydroponic system.展开更多
基金supported in part by the National Natural Science Foundation of China under grant nos.61573349,61703398 and 61673376in part by the National High Technology Research and Development Program of China(863 Program)under grant no.2015AA042308.
文摘In this study,a fast object detection algorithm based on binary deep convolution neural networks(CNNs)is proposed.Convolution kernels of different sizes are used to predict classes and bounding boxes of multi-scale objects directly in the last feature map of a deep CNN.In this way,rapid object detection with acceptable precision loss is achieved.In addition,binary quantisation for weight values and input data of each layer is used to squeeze the networks for faster object detection.Compared to full-precision convolution,the proposed binary deep CNNs for object detection results in 62 times faster convolutional operations and 32 times memory saving in theory,what’s more,the proposed method is easy to be implemented in embedded computing systems because of the binary operation for convolution and low memory requirement.Experimental results on Pascal VOC2007 validate the effectiveness of the authors’proposed method.
基金supported by the China Agriculture Research System of Ministry of Finance (MOF)and Ministry of Agriculture and Rural Affairs (MARA) (No.CARS-23-B15).
文摘In recent years,biochar has attracted considerable attention for soil quality improvement and carbon sequestration due to its unique physicochemical properties.However,the mechanism by which biochar application negatively affects the growth of crop seedlings has not been fully investigated.In this study,a hydroponic experiment was conducted to evaluate the response of rice,wheat,and corn seedlings to biochar application(CK,0 g/L;BC1,0.5 g/L;and BC2,1.0 g/L).Compared with the CK treatment,the BC1 and BC2 treatments decreased the fresh shoot and root weights of rice and corn seedlings(P<0.05),but there was no significant effect on wheat seedlings(P>0.05).For the contents of nutrient elements in seedlings,both BC1 and BC2 treatments hindered the roots from absorbing Fe and Cu and increased the uptake of Ca and Mn.Compared with the CK treatment,the translocation factor(TF)values of Ca,Mn,and Zn were significantly decreased especially in rice seedlings(35.3%-36.8%,68.7%-76.5%,and 29.8%-22.0%,respectively)under the BC1 and BC2 treatments,while only Mn was significantly decreased in wheat and corn seedlings(P<0.05).Transmission electron microscope(TEM)analysis of root cross-sections showed that nano-sized biochar particles(10~23 nm)were found in the root cells under BC2 treatment conditions.Our findings reveal that a large amount of biochar application can reduce nutrient absorption and translocation,and hinder rice,wheat,and corn seedlings,particularly rice seedling,in hydroponic system.