Asparagus stem blight is a devastating crop disease,and the early detection of its pathogenic spores is essential for effective disease control and prevention.However,spore detection is still hindered by complex backg...Asparagus stem blight is a devastating crop disease,and the early detection of its pathogenic spores is essential for effective disease control and prevention.However,spore detection is still hindered by complex backgrounds,small target sizes,and high annotation costs,which limit its practical application and widespread adoption.To address these issues,a semi-supervised spore detection framework is proposed for use under complex background conditions.Firstly,a difficulty perception scoring function is designed to quantify the detection difficulty of each image region.For regions with higher difficulty scores,a masking strategy is applied,while the remaining regions are adversarial augmentation is applied to encourage the model to learn fromchallenging areasmore effectively.Secondly,a Gaussian Mixture Model is employed to dynamically adjust the allocation threshold for pseudo-labels,thereby reducing the influence of unreliable supervision signals and enhancing the stability of semi-supervised learning.Finally,the Wasserstein distance is introduced for object localization refinement,offering a more robust positioning approach.Experimental results demonstrate that the proposed framework achieves 88.9% mAP50 and 60.7% mAP50-95,surpassing the baseline method by 4.2% and 4.6%,respectively,using only 10% of labeled data.In comparison with other state-of-the-art semi-supervised detection models,the proposed method exhibits superior detection accuracy and robustness.In conclusion,the framework not only offers an efficient and reliable solution for plant pathogen spore detection but also provides strong algorithmic support for real-time spore detection and early disease warning systems,with significant engineering application potential.展开更多
This study addresses the challenge posed by the small spore size of tomato gray mold,which hinders its identification and enumeration by conventional techniques.This work presents a novel approach for quantifying spor...This study addresses the challenge posed by the small spore size of tomato gray mold,which hinders its identification and enumeration by conventional techniques.This work presents a novel approach for quantifying spore counts of tomato gray mold using diffraction imaging technology and image processing techniques.To construct a device for acquiring diffraction images of tomato gray mold spores,initially,the hyperspectral data pertaining to the gray mold spores of tomatoes was obtained.The characteristic wavelength of the light source of the diffraction image acquisition device was obtained by smoothing,principal component analysis,and comprehensive coefficient weight calculation.Then,the key parameters of the system were simulated,and the diffraction image acquisition device was built.Finally,tomato gray mold spores were counted based on angular spectrum reconstruction and image processing.The findings indicated that the combined contribution rate of the initial and secondary principal components of the original spectral data obtained from tomato gray mold spore samples amounted to 92.271%.The visible range of 435 nm,475 nm,and 720 nm can be selected as the light source for tomato gray mold’s spore diffraction imaging system.CMOS image sensor was installed 45 mm below the micropore with a diameter of 100μm,and the diffraction image obtained by simulation has a clear diffraction fingerprint.The diffraction imaging system can collect diffraction images of disease spores,and the collected diffraction images have clear diffraction fingerprints.The experimental error range was 5.13%-8.57%,and the average error was 6.42%.The error was within a 95%consistency.Therefore,this study can provide a research basis for the classification and recognition of greenhouse disease spores.展开更多
Introduction:Raw milk is the basic raw material of dairy products.Bacillus cereus(B.cereus)is a typical conditional pathogenic bacteria and cold-phagocytic spoilage bacteria in raw milk.Materials and Methods:In this s...Introduction:Raw milk is the basic raw material of dairy products.Bacillus cereus(B.cereus)is a typical conditional pathogenic bacteria and cold-phagocytic spoilage bacteria in raw milk.Materials and Methods:In this study,a quantitative polymerase chain reaction(qPCR)method for detecting B.cereus in raw milk was established.The specificity of the method was verified by using other Bacillus bacteria and pathogenic bacteria;the sensitivity of the method was evaluated by preparing recombinant plasmids and simulated contaminated samples;and the applicability of the method was verified using pure spore DNA.The actual sample detection was completed by using the established qPCR method.Results:The qPCR established in this study can specifically detect B.cereus in raw milk.The limit of detection of the method was as low as 200 CFU/mL,the limit of quantification ranged from 2×10^(2)to 2×10^(8)CFU/mL,and the amplification efficiency of qPCR was 96.6%.Conclusions:The method established in this study can distinguish B.cereus from other Bacillus bacteria,and spore DNA can be used as the detection object.This method has the advantages of strong specificity,high sensitivity,wide application range,and short detection time,which isexpectedtobeapplied in thedairy industry.展开更多
基金supported by Development of asparagus price database based on agricultural big data(381724).
文摘Asparagus stem blight is a devastating crop disease,and the early detection of its pathogenic spores is essential for effective disease control and prevention.However,spore detection is still hindered by complex backgrounds,small target sizes,and high annotation costs,which limit its practical application and widespread adoption.To address these issues,a semi-supervised spore detection framework is proposed for use under complex background conditions.Firstly,a difficulty perception scoring function is designed to quantify the detection difficulty of each image region.For regions with higher difficulty scores,a masking strategy is applied,while the remaining regions are adversarial augmentation is applied to encourage the model to learn fromchallenging areasmore effectively.Secondly,a Gaussian Mixture Model is employed to dynamically adjust the allocation threshold for pseudo-labels,thereby reducing the influence of unreliable supervision signals and enhancing the stability of semi-supervised learning.Finally,the Wasserstein distance is introduced for object localization refinement,offering a more robust positioning approach.Experimental results demonstrate that the proposed framework achieves 88.9% mAP50 and 60.7% mAP50-95,surpassing the baseline method by 4.2% and 4.6%,respectively,using only 10% of labeled data.In comparison with other state-of-the-art semi-supervised detection models,the proposed method exhibits superior detection accuracy and robustness.In conclusion,the framework not only offers an efficient and reliable solution for plant pathogen spore detection but also provides strong algorithmic support for real-time spore detection and early disease warning systems,with significant engineering application potential.
基金partially supported by the National Natural Science Foundation of China(Grant Nos.32071905,3217895,and 32201686)Priority Academic Program Development of Jiangsu Higher Education Institutions(Grant No.PAPD-2023-87)+3 种基金Agricultural Equipment Department of Jiangsu University(Grant No.NZXB20210106)National Key Research and Development Program for Young Scientists(Grant No.2022YFD2000013)General Program of Basic Science(Natural Science)Research in Higher Education Institutions of Jiangsu Province(Grant No.23KJB210004)Natural Science Foundation of Jiangsu Province for Youth(Grant No.BK20240880).
文摘This study addresses the challenge posed by the small spore size of tomato gray mold,which hinders its identification and enumeration by conventional techniques.This work presents a novel approach for quantifying spore counts of tomato gray mold using diffraction imaging technology and image processing techniques.To construct a device for acquiring diffraction images of tomato gray mold spores,initially,the hyperspectral data pertaining to the gray mold spores of tomatoes was obtained.The characteristic wavelength of the light source of the diffraction image acquisition device was obtained by smoothing,principal component analysis,and comprehensive coefficient weight calculation.Then,the key parameters of the system were simulated,and the diffraction image acquisition device was built.Finally,tomato gray mold spores were counted based on angular spectrum reconstruction and image processing.The findings indicated that the combined contribution rate of the initial and secondary principal components of the original spectral data obtained from tomato gray mold spore samples amounted to 92.271%.The visible range of 435 nm,475 nm,and 720 nm can be selected as the light source for tomato gray mold’s spore diffraction imaging system.CMOS image sensor was installed 45 mm below the micropore with a diameter of 100μm,and the diffraction image obtained by simulation has a clear diffraction fingerprint.The diffraction imaging system can collect diffraction images of disease spores,and the collected diffraction images have clear diffraction fingerprints.The experimental error range was 5.13%-8.57%,and the average error was 6.42%.The error was within a 95%consistency.Therefore,this study can provide a research basis for the classification and recognition of greenhouse disease spores.
基金supported by the National Key Research and Development Program of China (2018YFC1604201 and 2018YFC1603800).
文摘Introduction:Raw milk is the basic raw material of dairy products.Bacillus cereus(B.cereus)is a typical conditional pathogenic bacteria and cold-phagocytic spoilage bacteria in raw milk.Materials and Methods:In this study,a quantitative polymerase chain reaction(qPCR)method for detecting B.cereus in raw milk was established.The specificity of the method was verified by using other Bacillus bacteria and pathogenic bacteria;the sensitivity of the method was evaluated by preparing recombinant plasmids and simulated contaminated samples;and the applicability of the method was verified using pure spore DNA.The actual sample detection was completed by using the established qPCR method.Results:The qPCR established in this study can specifically detect B.cereus in raw milk.The limit of detection of the method was as low as 200 CFU/mL,the limit of quantification ranged from 2×10^(2)to 2×10^(8)CFU/mL,and the amplification efficiency of qPCR was 96.6%.Conclusions:The method established in this study can distinguish B.cereus from other Bacillus bacteria,and spore DNA can be used as the detection object.This method has the advantages of strong specificity,high sensitivity,wide application range,and short detection time,which isexpectedtobeapplied in thedairy industry.
