Particulate matter（PM） from cooking has caused seriously indoor air pollutant and aroused risk to human health.It is urged to get deep knowledge of their spatial-temporal distribution of source emission characterist...Particulate matter（PM） from cooking has caused seriously indoor air pollutant and aroused risk to human health.It is urged to get deep knowledge of their spatial-temporal distribution of source emission characteristics,especially ultrafine particles（UFP ＆lt; 100 nm） and accumulation mode particles（AMP 100-555 nm）.Four commercial cooking oils are auto dipped water to simulate cooking fume under heating to 255℃ to investigate PM emission and decay features between 0.03 and 10 μm size dimension by electrical low pressure impactor（ELPI） without ventilation.Rapeseed and sunflower produced high PM_（2.5） around5.1 mg/m^3,in comparison with those of soybean and corn（5.87 and 4.55 mg/m^3,respectively）at peak emission time between 340 and 450 sec since heating oil,but with the same level of particle numbers 6-9 × 10~5/cm^3.Mean values of PM_（1.0）/PM_（2.5） and PM_（2.5）/PM_（10） at peak emission time are around 0.51-0.55 and 0.23-0.29.After 15 min naturally deposition,decay rates of PM_（1.0）,PM_（2.5） and PM_（10） are 13.3%-29.8%,20.1%-33.9%and 41.2%-54.7%,which manifest that PM_（1.0） is quite hard to decay than larger particles,PM_（2.5） and PM_（1.0）.The majority of the particle emission locates at 43 nm with the largest decay rate at 75%,and shifts to a larger size between137 and 555 nm after 15 min decay.The decay rates of the particles are sensitive to the oil type.展开更多
High-throughput phenotyping is the bottleneck for advancing field trait characterization and yield improvement in major field crops.Specifically for sorghum(Sorghum bicolor L.),rapid plant-level yield estimation is hi...High-throughput phenotyping is the bottleneck for advancing field trait characterization and yield improvement in major field crops.Specifically for sorghum(Sorghum bicolor L.),rapid plant-level yield estimation is highly dependent on characterizing the number of grains within a panicle.In this context,the integration of computer vision and artificial intelligence algorithms with traditional field phenotyping can be a critical solution to reduce labor costs and time.Therefore,this study aims to improve sorghum panicle detection and grain number estimation from smartphone-capture images under field conditions.A preharvest benchmark dataset was collected at field scale(2023 season,Kansas,USA),with 648 images of sorghum panicles retrieved via smartphone device,and grain number counted.Each sorghum panicle image was manually labeled,and the images were augmented.Two models were trained using the Detectron2 and Yolov8 frameworks for detection and segmentation,with an average precision of 75%and 89%,respectively.For the grain number,3 models were trained:MCNN(multiscale convolutional neural network),TCNN-Seed(two-column CNN-Seed),and Sorghum-Net(developed in this study).The Sorghum-Net model showed a mean absolute percentage error of 17%,surpassing the other models.Lastly,a simple equation was presented to relate the count from the model(using images from only one side of the panicle)to the field-derived observed number of grains per sorghum panicle.The resulting framework obtained an estimation of grain number with a 17%error.The proposed framework lays the foundation for the development of a more robust application to estimate sorghum yield using images from a smartphone at the plant level.展开更多
基金supported by the strategic project of science and technology of Chinese Academy of Sciences(No.XDB05050000)
文摘Particulate matter（PM） from cooking has caused seriously indoor air pollutant and aroused risk to human health.It is urged to get deep knowledge of their spatial-temporal distribution of source emission characteristics,especially ultrafine particles（UFP ＆lt; 100 nm） and accumulation mode particles（AMP 100-555 nm）.Four commercial cooking oils are auto dipped water to simulate cooking fume under heating to 255℃ to investigate PM emission and decay features between 0.03 and 10 μm size dimension by electrical low pressure impactor（ELPI） without ventilation.Rapeseed and sunflower produced high PM_（2.5） around5.1 mg/m^3,in comparison with those of soybean and corn（5.87 and 4.55 mg/m^3,respectively）at peak emission time between 340 and 450 sec since heating oil,but with the same level of particle numbers 6-9 × 10~5/cm^3.Mean values of PM_（1.0）/PM_（2.5） and PM_（2.5）/PM_（10） at peak emission time are around 0.51-0.55 and 0.23-0.29.After 15 min naturally deposition,decay rates of PM_（1.0）,PM_（2.5） and PM_（10） are 13.3%-29.8%,20.1%-33.9%and 41.2%-54.7%,which manifest that PM_（1.0） is quite hard to decay than larger particles,PM_（2.5） and PM_（1.0）.The majority of the particle emission locates at 43 nm with the largest decay rate at 75%,and shifts to a larger size between137 and 555 nm after 15 min decay.The decay rates of the particles are sensitive to the oil type.
文摘High-throughput phenotyping is the bottleneck for advancing field trait characterization and yield improvement in major field crops.Specifically for sorghum(Sorghum bicolor L.),rapid plant-level yield estimation is highly dependent on characterizing the number of grains within a panicle.In this context,the integration of computer vision and artificial intelligence algorithms with traditional field phenotyping can be a critical solution to reduce labor costs and time.Therefore,this study aims to improve sorghum panicle detection and grain number estimation from smartphone-capture images under field conditions.A preharvest benchmark dataset was collected at field scale(2023 season,Kansas,USA),with 648 images of sorghum panicles retrieved via smartphone device,and grain number counted.Each sorghum panicle image was manually labeled,and the images were augmented.Two models were trained using the Detectron2 and Yolov8 frameworks for detection and segmentation,with an average precision of 75%and 89%,respectively.For the grain number,3 models were trained:MCNN(multiscale convolutional neural network),TCNN-Seed(two-column CNN-Seed),and Sorghum-Net(developed in this study).The Sorghum-Net model showed a mean absolute percentage error of 17%,surpassing the other models.Lastly,a simple equation was presented to relate the count from the model(using images from only one side of the panicle)to the field-derived observed number of grains per sorghum panicle.The resulting framework obtained an estimation of grain number with a 17%error.The proposed framework lays the foundation for the development of a more robust application to estimate sorghum yield using images from a smartphone at the plant level.
