This study aimed to address the challenge of accurately and reliably detecting tomatoes in dense planting environments,a critical prerequisite for the automation implementation of robotic harvesting.However,the heavy ...This study aimed to address the challenge of accurately and reliably detecting tomatoes in dense planting environments,a critical prerequisite for the automation implementation of robotic harvesting.However,the heavy reliance on extensive manually annotated datasets for training deep learning models still poses significant limitations to their application in real-world agricultural production environments.To overcome these limitations,we employed domain adaptive learning approach combined with the YOLOv5 model to develop a novel tomato detection model called as TDA-YOLO(tomato detection domain adaptation).We designated the normal illumination scenes in dense planting environments as the source domain and utilized various other illumination scenes as the target domain.To construct bridge mechanism between source and target domains,neural preset for color style transfer is introduced to generate a pseudo-dataset,which served to deal with domain discrepancy.Furthermore,this study combines the semi-supervised learning method to enable the model to extract domain-invariant features more fully,and uses knowledge distillation to improve the model's ability to adapt to the target domain.Additionally,for purpose of promoting inference speed and low computational demand,the lightweight FasterNet network was integrated into the YOLOv5's C3 module,creating a modified C3_Faster module.The experimental results demonstrated that the proposed TDA-YOLO model significantly outperformed original YOLOv5s model,achieving a mAP(mean average precision)of 96.80%for tomato detection across diverse scenarios in dense planting environments,increasing by 7.19 percentage points;Compared with the latest YOLOv8 and YOLOv9,it is also 2.17 and 1.19 percentage points higher,respectively.The model's average detection time per image was an impressive 15 milliseconds,with a FLOPs(floating point operations per second)count of 13.8 G.After acceleration processing,the detection accuracy of the TDA-YOLO model on the Jetson Xavier NX development board is 90.95%,the mAP value is 91.35%,and the detection time of each image is 21 ms,which can still meet the requirements of real-time detection of tomatoes in dense planting environment.The experimental results show that the proposed TDA-YOLO model can accurately and quickly detect tomatoes in dense planting environment,and at the same time avoid the use of a large number of annotated data,which provides technical support for the development of automatic harvesting systems for tomatoes and other fruits.展开更多
In order to study the strength failure and crack coalescence characteristics of cracked rocks, uniaxial compression experiments were conducted on cylindrical sandstone specimens, sampled from Longyou Grottoes of Zheji...In order to study the strength failure and crack coalescence characteristics of cracked rocks, uniaxial compression experiments were conducted on cylindrical sandstone specimens, sampled from Longyou Grottoes of Zhejiang Province, China, with a single pre-cut crack soaking in different chemical solutions. Based on the results of uniaxial compressive test under different chemical solutions and velocities of flow, the effect of strength and deformation characteristics and main modes of crack coalescence for cracked rocks under chemical corrosion were analyzed. The results show that the pH value and velocity of the chemical solutions both have great influence on the sandstone sample's uniaxial compressive strength and deformation characteristics. Cracked sandstone samples are tension-destructed under uniaxial compression, and the crack propagation directions are consistent with the loading direction. The phenomena of crack initiation, propagation and coalescence of sandstone are well observed. Four different crack types are identified based on the crack propagation mechanism by analyzing the ultimate failure modes of sandstone containing a single pre-cut fissure. The failure process of specimen in air is similar with the specimen under chemical solutions, however, the initial time of crack occuring in specimen under chemical solutions is generally earlier than that in the natural specimen, and the crack propagation and coalescence process of specimen under chemical solutions are longer than those of the natural specimen due to softening of structure of rock caused by hydro-chemical action. Immersion velocity of flow and chemical solutions does not have influence on the ultimate modes of crack coalescence.展开更多
Despite the presence of a large area of andesite in the Sayaburi Province of Laos, it has received very little attention. Based on a combination of detailed field investigations, geochronology and geochemical analysis...Despite the presence of a large area of andesite in the Sayaburi Province of Laos, it has received very little attention. Based on a combination of detailed field investigations, geochronology and geochemical analysis, this study aims to explore the geochemical, Sr-Nd isotopic, and source rock characteristics, as well as the genesis and tectonic setting of the andesite in this region. In the Sayaburi Province, the andesite zircon U-Pb age is(241.2±1.2) Ma. The andesite rock is classified in the metaluminous-weak peraluminous calc-alkaline series. The light rare-earth elements(LREEs) are enriched and characterized by clear fractionation, whereas heavy rare-earth elements(HREEs) are relatively depleted and have no signs of fractionation. The average δEu is 0.96 with weak-or-no Eu anomalies. It is enriched in large ion lithophile elements such as Rb and K, while depleted in high field-strength elements such as Nb, Ta, P and Ti. For andesites in the Sayaburi Province, the(87Rb/86Sr)t value ranges in 0.702849-0.704687, the εNd(t) value is between 3.53 and 4.77, the tDM(t) value ranges in 633-835 Ma, and the tDM2(t) ranges in 625–724 Ma. The results based on the synthesis of petrology, geochemistry, and regional tectonic background studies show that 1) the andesitic magma source in the study area is an enriched mantle, which is modified by subduction zone fluids;2) the geotectonic background environment of the andesite in Sayaburi area is the continental island arc environment and related to the tectonic evolution of Jinghong–Nan–Uttaradit back-arc basin, which reflects that the magmatic source is enriched with a mantle wedge component modified by a subduction zone fluid(or melt).展开更多
The Lunggar iron deposit belongs to the Bangong-Nujiang metallogenic belt and is located in central Lhasa on the Tibetan Plateau.In the Lunggar deposit,iron mineralization formed in the skarnization contact zone betwe...The Lunggar iron deposit belongs to the Bangong-Nujiang metallogenic belt and is located in central Lhasa on the Tibetan Plateau.In the Lunggar deposit,iron mineralization formed in the skarnization contact zone between the Early Cretaceous granodiorite and the late Permian Xiala Formation limestone.In this study,we achieved detailed zircon U-Pb-Hf isotopes and mineral chemistry for the Early Cretaceous granodiorite.Zircon U-Pb dating results indicate that the Early Cretaceous granodiorite emplaced at ca.119 Ma.Based on the trace elements in zircons and the mineral chemical composition of amphibole and biotite,the Early Cretaceous granodiorite was believed to form under condition of high temperature(>700°C),low pressure(100400 MPa),and relatively high oxygen fugacity(lgfO2)(13.6 to 13.9)and H2O content(4%8%).Zircon trace elements,Hf isotope and biotite chemistry collectively reveal that significant juvenile mantle-derived magmas contributed to the source of the granodiorite.The relatively high logfO2 and shallow magma chamber are beneficial for skarn iron mineralization,implying remarkable potential for further prospecting in the Lunggar iron deposit.展开更多
基金The National Natural Science Foundation of China (32371993)The Natural Science Research Key Project of Anhui Provincial University(2022AH040125&2023AH040135)The Key Research and Development Plan of Anhui Province (202204c06020022&2023n06020057)。
文摘This study aimed to address the challenge of accurately and reliably detecting tomatoes in dense planting environments,a critical prerequisite for the automation implementation of robotic harvesting.However,the heavy reliance on extensive manually annotated datasets for training deep learning models still poses significant limitations to their application in real-world agricultural production environments.To overcome these limitations,we employed domain adaptive learning approach combined with the YOLOv5 model to develop a novel tomato detection model called as TDA-YOLO(tomato detection domain adaptation).We designated the normal illumination scenes in dense planting environments as the source domain and utilized various other illumination scenes as the target domain.To construct bridge mechanism between source and target domains,neural preset for color style transfer is introduced to generate a pseudo-dataset,which served to deal with domain discrepancy.Furthermore,this study combines the semi-supervised learning method to enable the model to extract domain-invariant features more fully,and uses knowledge distillation to improve the model's ability to adapt to the target domain.Additionally,for purpose of promoting inference speed and low computational demand,the lightweight FasterNet network was integrated into the YOLOv5's C3 module,creating a modified C3_Faster module.The experimental results demonstrated that the proposed TDA-YOLO model significantly outperformed original YOLOv5s model,achieving a mAP(mean average precision)of 96.80%for tomato detection across diverse scenarios in dense planting environments,increasing by 7.19 percentage points;Compared with the latest YOLOv8 and YOLOv9,it is also 2.17 and 1.19 percentage points higher,respectively.The model's average detection time per image was an impressive 15 milliseconds,with a FLOPs(floating point operations per second)count of 13.8 G.After acceleration processing,the detection accuracy of the TDA-YOLO model on the Jetson Xavier NX development board is 90.95%,the mAP value is 91.35%,and the detection time of each image is 21 ms,which can still meet the requirements of real-time detection of tomatoes in dense planting environment.The experimental results show that the proposed TDA-YOLO model can accurately and quickly detect tomatoes in dense planting environment,and at the same time avoid the use of a large number of annotated data,which provides technical support for the development of automatic harvesting systems for tomatoes and other fruits.
基金Projects(10472130,41202225) supported by the National Natural Science Foundation of China
文摘In order to study the strength failure and crack coalescence characteristics of cracked rocks, uniaxial compression experiments were conducted on cylindrical sandstone specimens, sampled from Longyou Grottoes of Zhejiang Province, China, with a single pre-cut crack soaking in different chemical solutions. Based on the results of uniaxial compressive test under different chemical solutions and velocities of flow, the effect of strength and deformation characteristics and main modes of crack coalescence for cracked rocks under chemical corrosion were analyzed. The results show that the pH value and velocity of the chemical solutions both have great influence on the sandstone sample's uniaxial compressive strength and deformation characteristics. Cracked sandstone samples are tension-destructed under uniaxial compression, and the crack propagation directions are consistent with the loading direction. The phenomena of crack initiation, propagation and coalescence of sandstone are well observed. Four different crack types are identified based on the crack propagation mechanism by analyzing the ultimate failure modes of sandstone containing a single pre-cut fissure. The failure process of specimen in air is similar with the specimen under chemical solutions, however, the initial time of crack occuring in specimen under chemical solutions is generally earlier than that in the natural specimen, and the crack propagation and coalescence process of specimen under chemical solutions are longer than those of the natural specimen due to softening of structure of rock caused by hydro-chemical action. Immersion velocity of flow and chemical solutions does not have influence on the ultimate modes of crack coalescence.
基金Projects(DD20160107,DD20150742)supported by the China Geological SurveyProject supported by the International Scientific Plan of the Qinghai-Xizang(Tibet)Plateau of Chengdu Center,China Geological Survey
文摘Despite the presence of a large area of andesite in the Sayaburi Province of Laos, it has received very little attention. Based on a combination of detailed field investigations, geochronology and geochemical analysis, this study aims to explore the geochemical, Sr-Nd isotopic, and source rock characteristics, as well as the genesis and tectonic setting of the andesite in this region. In the Sayaburi Province, the andesite zircon U-Pb age is(241.2±1.2) Ma. The andesite rock is classified in the metaluminous-weak peraluminous calc-alkaline series. The light rare-earth elements(LREEs) are enriched and characterized by clear fractionation, whereas heavy rare-earth elements(HREEs) are relatively depleted and have no signs of fractionation. The average δEu is 0.96 with weak-or-no Eu anomalies. It is enriched in large ion lithophile elements such as Rb and K, while depleted in high field-strength elements such as Nb, Ta, P and Ti. For andesites in the Sayaburi Province, the(87Rb/86Sr)t value ranges in 0.702849-0.704687, the εNd(t) value is between 3.53 and 4.77, the tDM(t) value ranges in 633-835 Ma, and the tDM2(t) ranges in 625–724 Ma. The results based on the synthesis of petrology, geochemistry, and regional tectonic background studies show that 1) the andesitic magma source in the study area is an enriched mantle, which is modified by subduction zone fluids;2) the geotectonic background environment of the andesite in Sayaburi area is the continental island arc environment and related to the tectonic evolution of Jinghong–Nan–Uttaradit back-arc basin, which reflects that the magmatic source is enriched with a mantle wedge component modified by a subduction zone fluid(or melt).
基金Project(2018YSJS14)supported by the Open Research Fund Program of Key Laboratory of Metallogenic Prediction of Nonferrous Metals and Geological Environment Monitoring(Central South University),Ministry of Education,China
文摘The Lunggar iron deposit belongs to the Bangong-Nujiang metallogenic belt and is located in central Lhasa on the Tibetan Plateau.In the Lunggar deposit,iron mineralization formed in the skarnization contact zone between the Early Cretaceous granodiorite and the late Permian Xiala Formation limestone.In this study,we achieved detailed zircon U-Pb-Hf isotopes and mineral chemistry for the Early Cretaceous granodiorite.Zircon U-Pb dating results indicate that the Early Cretaceous granodiorite emplaced at ca.119 Ma.Based on the trace elements in zircons and the mineral chemical composition of amphibole and biotite,the Early Cretaceous granodiorite was believed to form under condition of high temperature(>700°C),low pressure(100400 MPa),and relatively high oxygen fugacity(lgfO2)(13.6 to 13.9)and H2O content(4%8%).Zircon trace elements,Hf isotope and biotite chemistry collectively reveal that significant juvenile mantle-derived magmas contributed to the source of the granodiorite.The relatively high logfO2 and shallow magma chamber are beneficial for skarn iron mineralization,implying remarkable potential for further prospecting in the Lunggar iron deposit.
