Shijiuyang Constructed Wetland（110 hm^2） is a drinking water source treatment wetland with primary structural units of ponds and plant-bed/ditch systems. The wetland can process about 250,000 tonnes of source water ...Shijiuyang Constructed Wetland（110 hm^2） is a drinking water source treatment wetland with primary structural units of ponds and plant-bed/ditch systems. The wetland can process about 250,000 tonnes of source water in the Xincheng River every day and supplies raw water for Shijiuyang Drinking Water Plant. Daily data for 28 months indicated that the major water quality indexes of source water had been improved by one grade. The percentage increase for dissolved oxygen and the removal rates of ammonia nitrogen, iron and manganese were 73.63%, 38.86%, 35.64%, and 22.14% respectively. The treatment performance weight of ponds and plant-bed/ditch systems was roughly equal but they treated different pollutants preferentially. Most water quality indexes had better treatment efficacy with increasing temperature and inlet concentrations. These results revealed that the pond–wetland complexes exhibited strong buffering capacity for source water quality improvement. The treatment cost of Shijiuyang Drinking Water Plant was reduced by about 30.3%. Regional rainfall significantly determined the external river water levels and adversely deteriorated the inlet water quality, thus suggesting that the ＂hidden＂ diffuse pollution in the multitudinous stream branches as well as their catchments should be the controlling emphases for river source water protection in the future. The combination of pond and plant-bed/ditch systems provides a successful paradigm for drinking water source pretreatment. Three other drinking water source treatment wetlands with ponds and plant-bed/ditch systems are in operation or construction in the stream networks of the Yangtze River Delta and more people will be benefited.展开更多
Artificial neural networks(ANNs)have made great strides in the field of remote sensing image object detection.However,low detection efficiency and high power consumption have always been significant bottlenecks in rem...Artificial neural networks(ANNs)have made great strides in the field of remote sensing image object detection.However,low detection efficiency and high power consumption have always been significant bottlenecks in remote sensing.Spiking neural networks(SNNs)process information in the form of sparse spikes,creating the advantage of high energy efficiency for computer vision tasks.However,most studies have focused on simple classification tasks,and only a few researchers have applied SNNs to object detection in natural images.In this study,we consider the parsimonious nature of biological brains and propose a fast ANN-to-SNN conversion method for remote sensing image detection.We establish a fast sparse model for pulse sequence perception based on group sparse features and conduct transform-domain sparse resampling of the original images to enable fast perception of image features and encoded pulse sequences.In addition,to meet accuracy requirements in relevant remote sensing scenarios,we theoretically analyze the transformation error and propose channel self-decaying weighted normalization(CSWN)to eliminate neuron overactivation.We propose S3Det,a remote sensing image object detection model.Our experiments,based on a large publicly available remote sensing dataset,show that S3Det achieves an accuracy performance similar to that of the ANN.Meanwhile,our transformed network is only 24.32%as sparse as the benchmark and consumes only 1.46 W,which is 1/122 of the original algorithm’s power consumption.展开更多
基金upported by the National Natural Science Foundation of China (No. 51278487)the Major National Water Pollution Control and Management Project of China (Nos. 2012ZX07403-003-03, 2008ZX07421-001)+1 种基金the National Basic Research Program (973) of China (No. 2009CB421103)the Knowledge Innovation Program of the Chinese Academy of Sciences (No. KZCX2-EW-410-05)
文摘Shijiuyang Constructed Wetland（110 hm^2） is a drinking water source treatment wetland with primary structural units of ponds and plant-bed/ditch systems. The wetland can process about 250,000 tonnes of source water in the Xincheng River every day and supplies raw water for Shijiuyang Drinking Water Plant. Daily data for 28 months indicated that the major water quality indexes of source water had been improved by one grade. The percentage increase for dissolved oxygen and the removal rates of ammonia nitrogen, iron and manganese were 73.63%, 38.86%, 35.64%, and 22.14% respectively. The treatment performance weight of ponds and plant-bed/ditch systems was roughly equal but they treated different pollutants preferentially. Most water quality indexes had better treatment efficacy with increasing temperature and inlet concentrations. These results revealed that the pond–wetland complexes exhibited strong buffering capacity for source water quality improvement. The treatment cost of Shijiuyang Drinking Water Plant was reduced by about 30.3%. Regional rainfall significantly determined the external river water levels and adversely deteriorated the inlet water quality, thus suggesting that the ＂hidden＂ diffuse pollution in the multitudinous stream branches as well as their catchments should be the controlling emphases for river source water protection in the future. The combination of pond and plant-bed/ditch systems provides a successful paradigm for drinking water source pretreatment. Three other drinking water source treatment wetlands with ponds and plant-bed/ditch systems are in operation or construction in the stream networks of the Yangtze River Delta and more people will be benefited.
基金Project supported by the National Key Research and Development Program of China(No.2022YFB4500900)。
文摘Artificial neural networks(ANNs)have made great strides in the field of remote sensing image object detection.However,low detection efficiency and high power consumption have always been significant bottlenecks in remote sensing.Spiking neural networks(SNNs)process information in the form of sparse spikes,creating the advantage of high energy efficiency for computer vision tasks.However,most studies have focused on simple classification tasks,and only a few researchers have applied SNNs to object detection in natural images.In this study,we consider the parsimonious nature of biological brains and propose a fast ANN-to-SNN conversion method for remote sensing image detection.We establish a fast sparse model for pulse sequence perception based on group sparse features and conduct transform-domain sparse resampling of the original images to enable fast perception of image features and encoded pulse sequences.In addition,to meet accuracy requirements in relevant remote sensing scenarios,we theoretically analyze the transformation error and propose channel self-decaying weighted normalization(CSWN)to eliminate neuron overactivation.We propose S3Det,a remote sensing image object detection model.Our experiments,based on a large publicly available remote sensing dataset,show that S3Det achieves an accuracy performance similar to that of the ANN.Meanwhile,our transformed network is only 24.32%as sparse as the benchmark and consumes only 1.46 W,which is 1/122 of the original algorithm’s power consumption.
