This study develops a three-dimensional automated detection framework(PitScan)that systematically evaluates the severity and phenomenology of pitting corrosion.This framework uses a python-based algorithm to analyse m...This study develops a three-dimensional automated detection framework(PitScan)that systematically evaluates the severity and phenomenology of pitting corrosion.This framework uses a python-based algorithm to analyse microcomputer-tomography scans(μCT)of cylindrical specimens undergoing corrosion.The approach systematically identifies several surface-based corrosion features,enabling full spatial characterisation of pitting parameters,including pit density,pit size,pit depth as well as pitting factor according to ASTM G46-94.Furthermore,it is used to evaluate pitting formation in tensile specimens of a Rare Earth Magnesium alloy undergoing corrosion,and relationships between key pitting parameters and mechanical performance are established.Results demonstrated that several of the parameters described in ASTM G46-94,including pit number,pit density and pitting factor,showed little correlation to mechanical performance.However,this study did identify that other parameters showed strong correlations with the ultimate tensile strength and these tended to be directly linked to the reduction of the cross-sectional area of the specimen.Specifically,our results indicate,that parameters directly linked to the loss of the cross-sectional area(e.g.minimum material width),are parameters that are most suited to provide an indication of a specimen’s mechanical performance.The automated detection framework developed in this study has the potential to provide a basis to standardise measurements of pitting corrosion across a range of metals and future prediction of mechanical strength over degradation time.展开更多
Image segmentation directly determines the performance of automatic screening technique. However,there are overlapping nuclei in nuclei images. It raises a challenge to nuclei segmentation. To solve the problem,a segm...Image segmentation directly determines the performance of automatic screening technique. However,there are overlapping nuclei in nuclei images. It raises a challenge to nuclei segmentation. To solve the problem,a segmentation method of overlapping cervical nuclei based on the identification is proposed. This method consists of three stages： classifier training,recognition and fine segmentation. In the classifier training,feature selection and classifier selection are used to obtain a classifier with high recognition rate. In the recognition,the outputs of the rough segmentation are classified and processed according to their labels. In the fine segmentation,the severely overlapping nuclei are further segmented based on the prior knowledge provided by the recognition. Experiments show that this method can accurately segment overlapping nuclei.展开更多
基金funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 813869.
文摘This study develops a three-dimensional automated detection framework(PitScan)that systematically evaluates the severity and phenomenology of pitting corrosion.This framework uses a python-based algorithm to analyse microcomputer-tomography scans(μCT)of cylindrical specimens undergoing corrosion.The approach systematically identifies several surface-based corrosion features,enabling full spatial characterisation of pitting parameters,including pit density,pit size,pit depth as well as pitting factor according to ASTM G46-94.Furthermore,it is used to evaluate pitting formation in tensile specimens of a Rare Earth Magnesium alloy undergoing corrosion,and relationships between key pitting parameters and mechanical performance are established.Results demonstrated that several of the parameters described in ASTM G46-94,including pit number,pit density and pitting factor,showed little correlation to mechanical performance.However,this study did identify that other parameters showed strong correlations with the ultimate tensile strength and these tended to be directly linked to the reduction of the cross-sectional area of the specimen.Specifically,our results indicate,that parameters directly linked to the loss of the cross-sectional area(e.g.minimum material width),are parameters that are most suited to provide an indication of a specimen’s mechanical performance.The automated detection framework developed in this study has the potential to provide a basis to standardise measurements of pitting corrosion across a range of metals and future prediction of mechanical strength over degradation time.
基金supported by the National Natural Science Foundation of China(61673142,61471145,61305001)the Foundation of Education Department of Heilongjiang Province(12511096)+1 种基金the Research Fund for the Doctoral Program of Higher Education of China(20132303120003)the Science Funds for the Young Innovative Talents of HUST(20152)
文摘Image segmentation directly determines the performance of automatic screening technique. However,there are overlapping nuclei in nuclei images. It raises a challenge to nuclei segmentation. To solve the problem,a segmentation method of overlapping cervical nuclei based on the identification is proposed. This method consists of three stages： classifier training,recognition and fine segmentation. In the classifier training,feature selection and classifier selection are used to obtain a classifier with high recognition rate. In the recognition,the outputs of the rough segmentation are classified and processed according to their labels. In the fine segmentation,the severely overlapping nuclei are further segmented based on the prior knowledge provided by the recognition. Experiments show that this method can accurately segment overlapping nuclei.
