CN-121453831-B - Steel surface defect online detection method, device, system and storage medium

CN121453831BCN 121453831 BCN121453831 BCN 121453831BCN-121453831-B

Abstract

The invention provides a steel surface defect online detection method, device and system and a storage medium, and relates to the technical field of defect detection. The method comprises the steps of obtaining a preset reference spot map, wherein the reference spot map is obtained by establishing a coordinate system for a first aluminum nitride spot image of the surface of a defect-free steel sample under a scanning electron microscope, generating raised aluminum nitride spots on the surface of target steel by utilizing the residual temperature of the target steel after the target steel leaves an annealing furnace through a pulse nitriding process, collecting a second aluminum nitride spot image of the surface of the target steel, converting the second aluminum nitride spot image into the coordinate system, generating an online spot map, taking the reference spot map as priori knowledge of a pre-trained defect recognition model, and inputting the online spot map into the defect recognition model to recognize defects of the target steel. The invention can overcome the interference of the surface condition and the image quality of the steel, and improve the accuracy and the efficiency of online detection of the steel.

Inventors

HAO LIANG
ZHANG DI
ZHANG CONGCAN
AN BAO
KONG XIANGLING
XU YONG
LI SUFANG
ZHANG LEQI
FENG XIAOYANG
ZHANG WEN
WANG CONG

Assignees

河钢数字技术股份有限公司
雄安威赛博智能科技有限公司
哈尔滨工业大学(深圳)(哈尔滨工业大学深圳科技创新研究院)
河北科技大学

Dates

Publication Date: 20260508
Application Date: 20260106

Claims (8)

1. The online detection method for the surface defects of the steel is characterized by comprising the following steps of: obtaining a preset reference spot map, wherein the reference spot map is obtained by establishing a coordinate system for a first aluminum nitride spot image of the surface of a defect-free steel sample under a scanning electron microscope; after the target steel leaves the annealing furnace, utilizing the residual temperature of the target steel to generate raised aluminum nitride spots on the surface of the target steel through a pulse nitriding process; Collecting a second aluminum nitride spot image on the surface of the target steel, converting the second aluminum nitride spot image into the coordinate system, and generating an online spot map; Taking the reference spot map as priori knowledge of a pre-trained defect recognition model, inputting the online spot map into the defect recognition model, and recognizing defects of the target steel; The acquiring of the second aluminum nitride spot image of the surface of the target steel comprises the following steps: adopting a coaxial white light source to irradiate the surface of the target steel, and adopting a linear array camera system to scan the surface of the target steel; When the target steel moves for a certain distance, controlling the linear array camera system to synchronously acquire images of the surface of the target steel by adopting blue light and near infrared light, performing differential calculation on the blue light image and the near infrared light image, and reserving aluminum nitride spot signals to obtain the second aluminum nitride spot image; the converting the second aluminum nitride spot image into the coordinate system and generating an online spot map includes: acquiring a third aluminum nitride spot image of the surface of the first batch of off-line steel material in the current day under a scanning electron microscope in the coordinate system; converting the second aluminum nitride spot image into the coordinate system according to the third aluminum nitride spot image; and detecting position coordinates and size parameters of the aluminum nitride spots in the second aluminum nitride spot image under the coordinate system to obtain the online spot map.
2. The method according to claim 1, wherein the converting the second aluminum nitride spot image into the coordinate system based on the third aluminum nitride spot image comprises: respectively enhancing and extracting grain boundary contrast characteristic points of the third aluminum nitride spot image and the second aluminum nitride spot image according to a grain boundary network of the steel; And converting the second aluminum nitride spot image into the coordinate system according to the matching relation of the grain boundary contrast characteristic points of the third aluminum nitride spot image and the second aluminum nitride spot image.
3. The method for online detection of steel surface defects according to claim 1 or 2, wherein the acquiring a preset reference spot map comprises: establishing the coordinate system by taking the upper left corner of the first aluminum nitride spot image as an original point and taking pixels as unit lengths; And extracting position coordinates and size parameters of each aluminum nitride spot in the first aluminum nitride spot image according to a preset gray threshold value to obtain the preset reference spot map.
4. The method according to claim 1, wherein a penalty weight of defect recognition errors in the region in which AlN spots are identified in the reference spot map is set to be greater than 1 in a model loss function during pre-training of the defect recognition model.
5. The on-line detection method of steel surface defects according to claim 1, wherein the generating raised aluminum nitride spots on the surface of the target steel by a pulse nitriding process comprises: pulse spraying is carried out on the target steel by controlling high-purity ammonia through an electromagnetic valve, so as to generate active nitrogen atoms; And combining the active nitrogen atoms with aluminum elements in the target steel, and generating aluminum nitride spots which are raised on the surface of the target steel and firmly attached to the substrate of the target steel in situ.
6. An on-line detection device for steel surface defects, which is characterized by comprising: the reference spot map is obtained by establishing a coordinate system for a first aluminum nitride spot image on the surface of a non-defective steel sample under a scanning electron microscope; the surface spot marking module is used for generating raised aluminum nitride spots on the surface of the target steel through a pulse nitriding process by utilizing the residual temperature of the target steel after the target steel leaves an annealing furnace; The online map generation module is used for acquiring a second aluminum nitride spot image of the surface of the target steel, converting the second aluminum nitride spot image into the coordinate system and generating an online spot map, adopting a coaxial white light source to irradiate the surface of the target steel, adopting a linear array camera system to scan the surface of the target steel, controlling the linear array camera system to synchronously acquire the image of the surface of the target steel by adopting blue light and near infrared light each time the target steel moves for a certain distance, carrying out differential calculation on the blue light image and the near infrared light image, retaining aluminum nitride spot signals to obtain a second aluminum nitride spot image, acquiring a third aluminum nitride spot image of the surface of the first batch of offline steel in the coordinate system under a scanning electron microscope, converting the second aluminum nitride spot image into the coordinate system according to the third aluminum nitride spot image, and detecting position coordinates and size parameters of aluminum nitride spots in the second aluminum nitride spot image under the coordinate system to obtain the online spot map; And the defect detection module is used for taking the reference spot map as priori knowledge of a pre-trained defect identification model, inputting the online spot map into the defect identification model and identifying the defects of the target steel.
7. An on-line detection system for steel surface defects, characterized by comprising a memory and a processor, the memory storing a computer program, the processor implementing the method according to any one of claims 1 to 5 when executing the computer program.
8. A computer readable storage medium, characterized in that the computer readable storage medium stores a computer program which, when executed by a processor, implements the method according to any one of claims 1 to 5.

Description

Steel surface defect online detection method, device, system and storage medium Technical Field The present invention relates to the field of defect detection technologies, and in particular, to a method, an apparatus, a system, and a storage medium for online detection of a steel surface defect. Background The surface quality of the steel directly determines the mechanical property and service life of the steel, and particularly for special steel such as electrical steel, the core performance of the steel can be seriously affected by surface defects. Along with the development of high-speed and continuous steel production, a high-efficiency and accurate surface defect online detection technology has become a core requirement for quality control. At present, the surface defect detection of the steel is mainly performed by adopting manual visual inspection or traditional machine vision inspection, the manual inspection is highly dependent on experience judgment, and the traditional machine vision inspection is based on a basic defect morphology recognition algorithm, so that the application of the non-contact characteristic in the electrical steel detection is wider. However, the traditional machine vision detection is easily affected by the surface conditions such as oil films and the quality of collected images, the precision of identifying the micro defects is not enough, meanwhile, a large number of samples are relied on, and stable and reliable identification is difficult to realize when the defect samples are scarce. Therefore, there is a need for an online detection method for steel surface defects, which can overcome the surface condition and image quality interference of the steel, so as to improve the online detection accuracy and efficiency of the steel. Disclosure of Invention In view of the above, the embodiments of the present invention provide a method, an apparatus, a system, and a storage medium for online detection of surface defects of steel, so as to improve the accuracy and efficiency of online detection of surface defects of steel. In a first aspect, an embodiment of the present invention provides a method for online detecting a surface defect of a steel product, including: obtaining a preset reference spot map, wherein the reference spot map is obtained by establishing a coordinate system for a first aluminum nitride spot image of the surface of a defect-free steel sample under a scanning electron microscope; after the target steel leaves the annealing furnace, utilizing the residual temperature of the target steel to generate raised aluminum nitride spots on the surface of the target steel through a pulse nitriding process; Collecting a second aluminum nitride spot image on the surface of the target steel, converting the second aluminum nitride spot image into the coordinate system, and generating an online spot map; and taking the reference spot map as priori knowledge of a pre-trained defect recognition model, inputting the online spot map into the defect recognition model, and recognizing the defects of the target steel. In one possible implementation manner, the acquiring the second aluminum nitride spot image of the target steel surface includes: adopting a coaxial white light source to irradiate the surface of the target steel, and adopting a linear array camera system to scan the surface of the target steel; And when the target steel moves for a certain distance, controlling the linear array camera system to synchronously acquire images of the surface of the target steel by adopting blue light and near infrared light, performing differential calculation on the blue light image and the near infrared light image, and reserving aluminum nitride spot signals to obtain the second aluminum nitride spot image. In one possible implementation, the converting the second aluminum nitride spot image into the coordinate system and generating an online spot map includes: acquiring a third aluminum nitride spot image of the surface of the first batch of off-line steel material in the current day under a scanning electron microscope in the coordinate system; converting the second aluminum nitride spot image into the coordinate system according to the third aluminum nitride spot image; and detecting position coordinates and size parameters of the aluminum nitride spots in the second aluminum nitride spot image under the coordinate system to obtain the online spot map. In one possible implementation, the converting the second aluminum nitride spot image into the coordinate system according to the third aluminum nitride spot image includes: respectively enhancing and extracting grain boundary contrast characteristic points of the third aluminum nitride spot image and the second aluminum nitride spot image according to a grain boundary network of the steel; And converting the second aluminum nitride spot image into the coordinate system according to the matching relation of the grain boundary cont