Growth-related traits,such as aboveground biomass and leaf area,are critical indicators to characterize the growth of greenhouse lettuce.Currently,nondestructive methods for estimating growth-related traits are subjec...Growth-related traits,such as aboveground biomass and leaf area,are critical indicators to characterize the growth of greenhouse lettuce.Currently,nondestructive methods for estimating growth-related traits are subject to limitations in that the methods are susceptible to noise and heavily rely on manually designed features.In this study,a method for monitoring the growth of greenhouse lettuce was proposed by using digital images and a convolutional neural network(CNN).Taking lettuce images as the input,a CNN model was trained to learn the relationship between images and the corresponding growth-related traits,i.e.,leaf fresh weight(LFW),leaf dry weight(LDW),and leaf area(LA).To compare the results of the CNN model,widely adopted methods were also used.The results showed that the values estimated by CNN had good agreement with the actual measurements,with R^(2) values of 0.8938,0.8910,and 0.9156 and normalized root mean square error(NRMSE)values of 26.00,22.07,and 19.94%,outperforming the compared methods for all three growth-related traits.The obtained results showed that the CNN demonstrated superior estimation performance for the flat-type cultivars of Flandria and Tiberius compared with the curled-type cultivar of Locarno.Generalization tests were conducted by using images of Tiberius from another growing season.The results showed that the CNN was still capable of achieving accurate estimation of the growth-related traits,with R2 values of 0.9277,0.9126,and 0.9251 and NRMSE values of 22.96,37.29,and 27.60%.The results indicated that a CNN with digital images is a robust tool for the monitoring of the growth of greenhouse lettuce.展开更多
Rapid and accurate detection of pathogen spores is an important step to achieve early diagnosis of diseases in precision agriculture.Traditional detection methods are time-consuming,laborious,and subjective,and image ...Rapid and accurate detection of pathogen spores is an important step to achieve early diagnosis of diseases in precision agriculture.Traditional detection methods are time-consuming,laborious,and subjective,and image processing methods mainly rely on manually designed features that are difficult to cope with pathogen spore detection in complex scenes.Therefore,an MG-YOLO detection algorithm(Multi-head self-attention and Ghost-optimized YOLO)is proposed to detect gray mold spores rapidly.Firstly,Multi-head self-attention is introduced in the backbone to capture the global information of the pathogen spores.Secondly,we combine weighted Bidirectional Feature Pyramid Network(BiFPN)to fuse multiscale features of different layers.Then,a lightweight network is used to construct GhostCSP to optimize the neck part.Cucumber gray mold spores are used as the study object.The experimental results show that the improved MG-YOLO model achieves an accuracy of 0.983 for detecting gray mold spores and takes 0.009 s per image,which is significantly better than the state-of-the-art model.The visualization of the detection results shows that MG-YOLO effectively solves the detection of spores in blurred,small targets,multimorphology,and high-density scenes.Meanwhile,compared with the YOLOv5 model,the detection accuracy of the improved model is improved by 6.8%.It can meet the demand for high-precision detection of spores and provides a novel method to enhance the objectivity of pathogen spore detection.展开更多
Uneven illumination and clutter background were the most challenging problems to segmentation of disease symptom images.In order to achieve robust segmentation,a method for processing greenhouse vegetable foliar disea...Uneven illumination and clutter background were the most challenging problems to segmentation of disease symptom images.In order to achieve robust segmentation,a method for processing greenhouse vegetable foliar disease symptom images was proposed in this paper.The segmentation method was based on a decision tree which was constructed by a two-step coarse-to-fine procedure.Firstly,a coarse decision tree was built by the CART(Classification and Regression Tree)algorithm with a feature subset.The feature subset consisted of color features that was selected by Pearson’s Rank correlations.Then,the coarse decision tree was optimized by pruning.Using the optimized decision tree,segmentation of disease symptom images was achieved by conducting pixel-wise classification.In order to evaluate the robustness and accuracy of the proposed method,an experiment was performed using greenhouse cucumber downy mildew images.Results showed that the proposed method achieved an overall accuracy of 90.67%,indicating that the method was able to obtain robust segmentation of disease symptom images.展开更多
A new type of roof structure was developed for the shade room in a double-slope greenhouse used for mushroom-vegetable planting.A simulation model was developed to evaluate the thermal performance of the new roof with...A new type of roof structure was developed for the shade room in a double-slope greenhouse used for mushroom-vegetable planting.A simulation model was developed to evaluate the thermal performance of the new roof with an insulation thickness of 0.12 m in Beijing,China.The results showed that(1)the indoor air temperature of the shade room with the newly implemented shade roof was 2.7℃-4.9℃ higher than that of an ordinary shade room during the winter months;(2)The indoor air temperature of the solar room adjacent to the shade room with the new roof was higher than that of the ordinary solar room and the minimum indoor air temperature of the solar room was increased 1.9℃ at winter night;(3)the indoor temperature of the shade room with the new roof design was 2℃-4℃ lower than that of the ordinary shade room during the summer months;(4)Under factory production conditions,which were conducted in a controlled environment to promote the annual growth of the edible fungus,the heating energy consumption of the shade room after the implementation of the new roof structure was reduced by 69.3%,the amounted to total energy savings of 61.3% per year.The new roof structure provided a significant improvement in the thermal environment compared to an ordinary shade room,improved the vegetable growth in the winter,and also significantly reduced the energy consumption and production costs.展开更多
Travel recommendations form a major part of tourism service. Traditional collaborative filtering and Markov model are not appropriate for expressing the trajectory features,for travel preferences of tourists are dynam...Travel recommendations form a major part of tourism service. Traditional collaborative filtering and Markov model are not appropriate for expressing the trajectory features,for travel preferences of tourists are dynamic and affected by previous behaviors. Inspired by the success of deep learning in sequence learning,a personalized recurrent neural network (P-RecN) is proposed for tourist route recommendation. It is data-driven and adaptively learns the unknown mapping of historical trajectory input to recommended route output. Specifically,a trajectory encoding module is designed to mine the semantic information of trajectory data,and LSTM neural networks are used to capture the sequence travel patterns of tourists. In particular,a temporal attention mechanism is integrated to emphasize the main behavioral intention of tourists. We retrieve a geotagged photo dataset in Shanghai,and evaluate our model in terms of accuracy and ranking ability. Experimental results illustrated that P-RecN outperforms other baseline approaches and can effectively understand the travel patterns of tourists.展开更多
China is the largest producer and consumer of vegetables,its vegetable industry is playing an important role in the domestic agricultural sector and global vegetable export market.It is important to promote the long-t...China is the largest producer and consumer of vegetables,its vegetable industry is playing an important role in the domestic agricultural sector and global vegetable export market.It is important to promote the long-term sustainable development of Chinese vegetable industry for food security and quality of people’s lives.To find out the intrinsic way to promote the development of Chinese vegetable industry,this paper analyzed the influencing factors of Chinese vegetable production by utilizing the LMDI method and demonstrated the spatial-temporal characteristics of vegetable production through application of the Arc-GIS spatial autocorrelation analysis method.The results showed that the influencing factors of vegetable production were the cultivated land area,multiple cropping index,vegetable planting proportion and vegetable yield per hectare in China.The major driving factor had changed from vegetable planting proportion to vegetable yield per hectare.The influencing degrees of factors on vegetable production are different in different regions,regionalization is therefore a major feature of Chinese vegetable production.The government should take production technology,regionalization-driven effect,and marketing integration into consideration to promote the development of Chinese vegetable industry.展开更多
With the increasing demand for food worldwide,it has attracted increasing attention how to improve the agricultural production efficiency.This paper aims to develop a technical efficiency evaluation system for vegetab...With the increasing demand for food worldwide,it has attracted increasing attention how to improve the agricultural production efficiency.This paper aims to develop a technical efficiency evaluation system for vegetable production to provided decisions for the practice of precision agriculture.The paper analyses the system-needs and business processes,and proposes a system framework which has three tiers architectures,based on B/S model.The stochastic frontier analysis(SFA)algorithm model which is the incorporated into the system is established.The system was tested and evaluated by real business data,which were from Beijing from 2003 to 2011 to test system performance based on the temporal perspective and China during 2011 and 2012 to test system performance based on the spatial characteristics.The results shows that the system achieves the business requirements with an intelligent tool for data management and technical efficiency evaluation for vegetable production to improve automation,efficiency and convenience.展开更多
基金supported by the Beijing Leafy Vegetables Innovation Team of Modern Agro-industry Technology Research System(BAIC07-2020)the National Key Research and Development Project of Shandong(2017CXGC0201).
文摘Growth-related traits,such as aboveground biomass and leaf area,are critical indicators to characterize the growth of greenhouse lettuce.Currently,nondestructive methods for estimating growth-related traits are subject to limitations in that the methods are susceptible to noise and heavily rely on manually designed features.In this study,a method for monitoring the growth of greenhouse lettuce was proposed by using digital images and a convolutional neural network(CNN).Taking lettuce images as the input,a CNN model was trained to learn the relationship between images and the corresponding growth-related traits,i.e.,leaf fresh weight(LFW),leaf dry weight(LDW),and leaf area(LA).To compare the results of the CNN model,widely adopted methods were also used.The results showed that the values estimated by CNN had good agreement with the actual measurements,with R^(2) values of 0.8938,0.8910,and 0.9156 and normalized root mean square error(NRMSE)values of 26.00,22.07,and 19.94%,outperforming the compared methods for all three growth-related traits.The obtained results showed that the CNN demonstrated superior estimation performance for the flat-type cultivars of Flandria and Tiberius compared with the curled-type cultivar of Locarno.Generalization tests were conducted by using images of Tiberius from another growing season.The results showed that the CNN was still capable of achieving accurate estimation of the growth-related traits,with R2 values of 0.9277,0.9126,and 0.9251 and NRMSE values of 22.96,37.29,and 27.60%.The results indicated that a CNN with digital images is a robust tool for the monitoring of the growth of greenhouse lettuce.
基金the financial support of the National Natural Science Foundation of China(no.62176261).
文摘Rapid and accurate detection of pathogen spores is an important step to achieve early diagnosis of diseases in precision agriculture.Traditional detection methods are time-consuming,laborious,and subjective,and image processing methods mainly rely on manually designed features that are difficult to cope with pathogen spore detection in complex scenes.Therefore,an MG-YOLO detection algorithm(Multi-head self-attention and Ghost-optimized YOLO)is proposed to detect gray mold spores rapidly.Firstly,Multi-head self-attention is introduced in the backbone to capture the global information of the pathogen spores.Secondly,we combine weighted Bidirectional Feature Pyramid Network(BiFPN)to fuse multiscale features of different layers.Then,a lightweight network is used to construct GhostCSP to optimize the neck part.Cucumber gray mold spores are used as the study object.The experimental results show that the improved MG-YOLO model achieves an accuracy of 0.983 for detecting gray mold spores and takes 0.009 s per image,which is significantly better than the state-of-the-art model.The visualization of the detection results shows that MG-YOLO effectively solves the detection of spores in blurred,small targets,multimorphology,and high-density scenes.Meanwhile,compared with the YOLOv5 model,the detection accuracy of the improved model is improved by 6.8%.It can meet the demand for high-precision detection of spores and provides a novel method to enhance the objectivity of pathogen spore detection.
基金The authors would like to thank the financial support provided by The National Key Research and Development Program of China(2016YFD0300606,2017YFD0300402 and 2017YFD0300401).
文摘Uneven illumination and clutter background were the most challenging problems to segmentation of disease symptom images.In order to achieve robust segmentation,a method for processing greenhouse vegetable foliar disease symptom images was proposed in this paper.The segmentation method was based on a decision tree which was constructed by a two-step coarse-to-fine procedure.Firstly,a coarse decision tree was built by the CART(Classification and Regression Tree)algorithm with a feature subset.The feature subset consisted of color features that was selected by Pearson’s Rank correlations.Then,the coarse decision tree was optimized by pruning.Using the optimized decision tree,segmentation of disease symptom images was achieved by conducting pixel-wise classification.In order to evaluate the robustness and accuracy of the proposed method,an experiment was performed using greenhouse cucumber downy mildew images.Results showed that the proposed method achieved an overall accuracy of 90.67%,indicating that the method was able to obtain robust segmentation of disease symptom images.
基金This research was made possible through financial support from the Beijing Leafy Vegetables Innovation Team of Modern Agro-industry Technology Research System(BAIC07-2019)Yantai Science and Technology Development Project(2013ZH083).
文摘A new type of roof structure was developed for the shade room in a double-slope greenhouse used for mushroom-vegetable planting.A simulation model was developed to evaluate the thermal performance of the new roof with an insulation thickness of 0.12 m in Beijing,China.The results showed that(1)the indoor air temperature of the shade room with the newly implemented shade roof was 2.7℃-4.9℃ higher than that of an ordinary shade room during the winter months;(2)The indoor air temperature of the solar room adjacent to the shade room with the new roof was higher than that of the ordinary solar room and the minimum indoor air temperature of the solar room was increased 1.9℃ at winter night;(3)the indoor temperature of the shade room with the new roof design was 2℃-4℃ lower than that of the ordinary shade room during the summer months;(4)Under factory production conditions,which were conducted in a controlled environment to promote the annual growth of the edible fungus,the heating energy consumption of the shade room after the implementation of the new roof structure was reduced by 69.3%,the amounted to total energy savings of 61.3% per year.The new roof structure provided a significant improvement in the thermal environment compared to an ordinary shade room,improved the vegetable growth in the winter,and also significantly reduced the energy consumption and production costs.
基金supported in part by the National Natural Science Foundation of China (42171460)the Open Fund of Henan Key Laboratory for Synergistic Prevention of Water and Soil Environmental Pollution,Xinyang Normal University (KLSPWSEP-A09).
文摘Travel recommendations form a major part of tourism service. Traditional collaborative filtering and Markov model are not appropriate for expressing the trajectory features,for travel preferences of tourists are dynamic and affected by previous behaviors. Inspired by the success of deep learning in sequence learning,a personalized recurrent neural network (P-RecN) is proposed for tourist route recommendation. It is data-driven and adaptively learns the unknown mapping of historical trajectory input to recommended route output. Specifically,a trajectory encoding module is designed to mine the semantic information of trajectory data,and LSTM neural networks are used to capture the sequence travel patterns of tourists. In particular,a temporal attention mechanism is integrated to emphasize the main behavioral intention of tourists. We retrieve a geotagged photo dataset in Shanghai,and evaluate our model in terms of accuracy and ranking ability. Experimental results illustrated that P-RecN outperforms other baseline approaches and can effectively understand the travel patterns of tourists.
基金the financial support from Beijing Social Science Foundation(16YJA007)the earmarked fund for Beijing Innovation Consortium of Agriculture Research System(BAIC07-2013).
文摘China is the largest producer and consumer of vegetables,its vegetable industry is playing an important role in the domestic agricultural sector and global vegetable export market.It is important to promote the long-term sustainable development of Chinese vegetable industry for food security and quality of people’s lives.To find out the intrinsic way to promote the development of Chinese vegetable industry,this paper analyzed the influencing factors of Chinese vegetable production by utilizing the LMDI method and demonstrated the spatial-temporal characteristics of vegetable production through application of the Arc-GIS spatial autocorrelation analysis method.The results showed that the influencing factors of vegetable production were the cultivated land area,multiple cropping index,vegetable planting proportion and vegetable yield per hectare in China.The major driving factor had changed from vegetable planting proportion to vegetable yield per hectare.The influencing degrees of factors on vegetable production are different in different regions,regionalization is therefore a major feature of Chinese vegetable production.The government should take production technology,regionalization-driven effect,and marketing integration into consideration to promote the development of Chinese vegetable industry.
基金support from Beijing Social Science Foundation(16YJA007)the earmarked fund for Beijing Innovation Consortium of Agriculture Research System(BAIC07-20).
文摘With the increasing demand for food worldwide,it has attracted increasing attention how to improve the agricultural production efficiency.This paper aims to develop a technical efficiency evaluation system for vegetable production to provided decisions for the practice of precision agriculture.The paper analyses the system-needs and business processes,and proposes a system framework which has three tiers architectures,based on B/S model.The stochastic frontier analysis(SFA)algorithm model which is the incorporated into the system is established.The system was tested and evaluated by real business data,which were from Beijing from 2003 to 2011 to test system performance based on the temporal perspective and China during 2011 and 2012 to test system performance based on the spatial characteristics.The results shows that the system achieves the business requirements with an intelligent tool for data management and technical efficiency evaluation for vegetable production to improve automation,efficiency and convenience.
