KR-20260064160-A - Vision inspection equipment and the method for surface defection inspection

KR20260064160AKR 20260064160 AKR20260064160 AKR 20260064160AKR-20260064160-A

Abstract

The present invention relates to a vision inspection device for detecting surface defects. The present invention comprises: a detection server that receives data for identifying surface defects regarding a product to be inspected being transported via a transport means; a transport means driving unit that automatically transports the product to be inspected at a set speed; a surface detection unit that receives an acoustic signal generated by an ultrasound transmitted to the surface of the product to be inspected being transported and generates an image; a shooting device that recognizes the external shape of the product to be inspected being transported by the transport means and a barcode or QR code printed on the outside; a surface defect detection model that receives data converted into a 3D image from the image data generated from the shooting device within the detection server and transmits it to a terminal along with digitized data of each product to be inspected; an information processing device that displays a 3D image and numerical values of the product to be inspected, distinguishes and displays surface defects, and transmits them to a smartphone; an interface that inputs the shape of the product to be inspected and the allowable range of surface curvature through the detection server; and a surface defect product collection unit that separately collects and processes the product in which the surface defect has been confirmed. It is characterized by including a smartphone that transmits information on the occurrence of surface defects of the product subject to inspection to a manager.

Inventors

김종열
전주진

Assignees

주식회사 오리엔트

Dates

Publication Date: 20260507
Application Date: 20241031

Claims (8)

In a vision inspection device for detecting surface defects, A detection server that receives data for identifying surface defects in a product to be inspected being transported via a transport means; A conveying means that automatically conveys the above-mentioned product to be inspected at a set speed through a conveying means driving unit; A surface detection device that generates an image by receiving an acoustic signal generated by an ultrasound transmitted to the surface of the above-mentioned inspection target product being transported; An image capturing device that recognizes the external shape of the product to be inspected and a barcode or QR code printed on the outside, which is transported by the above-mentioned transport means; A surface defect detection model that receives data converted into a 3D image from image capturing device within the detection server and transmits it to an information processing device together with data of each product to be inspected; An information processing device that displays a 3D image and numerical values of the above-mentioned product subject to inspection, distinguishes and marks surface defects, and transmits them to a smartphone; An interface for inputting the shape and allowable range of surface curvature of the product to be inspected through the detection server; A surface defect product collection unit that separately collects and processes products in which the above-mentioned surface defects have been confirmed; and A vision inspection device for detecting surface defects characterized by including a smartphone that transmits information on the occurrence of surface defects of the product subject to inspection to an administrator.
In paragraph 1, A vision inspection device for detecting surface defects, characterized in that the information processing device has a database (DB) constructed to store inspection result data, including the production date, inspection type, and type of surface defect occurrence of the product to be inspected, from the image obtained from the image capturing device, in an inspection data storage device.
In paragraph 1, A vision inspection device for detecting surface defects, characterized by detecting whether each product subject to inspection has a surface defect through a surface defect detection model for products that are automatically moved and shipped by the above-mentioned conveying means, and detecting products subject to inspection that are confirmed to have surface defects before shipment by using a barcode or QR code attached to the surface of the product subject to inspection.
In paragraph 1, A vision inspection device for detecting surface defects, characterized by further including a deformation detection model that compares external shape image data generated by the above image capturing device with external shape image data, which is learning data obtained through learning based on the deformation of the product to be inspected, to determine whether each product to be inspected has a surface defect, wherein, if a surface defect is determined, the surface defect detection process of the above surface defect detection model is skipped and transmitted to the above information processing device along with a 3D image to display the product with the surface defect as a 3D image.
In paragraph 1, A vision inspection device for detecting surface defects, characterized in that the transfer speed of the above-mentioned transfer means is determined by considering the detection performance of the detection server and the detection time based on the image recognition speed of the image capturing device.
In paragraph 4, A vision inspection device for detecting surface defects, characterized by further including an output unit that outputs a warning light and determines a surface defect when the curvature exceeds the allowable range according to the determination of the surface defect detection model.
In a vision inspection method for detecting surface defects, A step in which data for detecting surface defects of a product to be inspected is input to a detection server through an interface with an information processing device regarding the product to be inspected; A step of inputting the transfer speed of a transfer means for transferring the above-mentioned product to be inspected; A step in which the product to be inspected is transported through the above-mentioned transport means; A step of receiving an acoustic signal generated by an ultrasound transmitted to the above-mentioned product subject to inspection and generating an acoustic signal image; A step of image recognition of a barcode or QR code printed on the exterior of each of the above-mentioned inspection target products and an external shape; A step of automatically quantifying the surface condition of each product subject to inspection and determining whether there is a surface defect by means of a surface defect detection model that compares image data of sound wave signals according to each product subject to inspection with image data of sound wave signals, which is training data according to the surface condition of the product subject to inspection; A step of transmitting a 3D image already generated within the detection server and quantified surface defects of each product to be inspected to a terminal via the surface defect detection model; A step of displaying the numerical values of each product subject to inspection and the 3D image on the terminal; A vision inspection method for detecting surface defects, characterized by including the step of verifying the optical code for the inspection target product in which the surface defect is confirmed, and processing and performing it as a separate item.
As a computer program stored on a non-transitory computer-readable storage medium, The above computer program includes one or more instructions, and when the instructions are executed by a computing device having one or more processors, the computing device, A step in which data for detecting surface defects of a product to be inspected is input to a detection server through an interface with an information processing device regarding the product to be inspected; A step of inputting the transfer speed of a transfer means for transferring the above-mentioned product to be inspected; A step in which the product to be inspected is transported through the above-mentioned transport means; A step of receiving an acoustic signal generated by an ultrasound transmitted to the above-mentioned product subject to inspection and generating an acoustic signal image; A step of image recognition of a barcode or QR code printed on the exterior of each of the above-mentioned inspection target products and an external shape; A step of automatically quantifying the surface condition of each product subject to inspection and determining whether there is a surface defect by means of a surface defect detection model that compares image data of sound wave signals according to each product subject to inspection with image data of sound wave signals, which is training data according to the surface condition of the product subject to inspection; A step of transmitting a 3D image already generated within the detection server and quantified surface defects of each product to be inspected to a terminal via the surface defect detection model; A step of displaying the numerical values of each product subject to inspection and the 3D image on the terminal; A computer program that performs the step of verifying the optical code for the inspection target product with the above-mentioned surface defect, and processing and performing it as a separate item.

Description

Vision inspection equipment and the method for surface defect detection The present invention relates to a vision inspection device and method for detecting surface defects, and more specifically, to a vision inspection device and method for detecting surface defects that can improve productivity by preventing products with surface defects in advance and collecting them separately. This achievement was carried out with the support of the "Dongmyung University Innovation Research Complex Development Project" under the University Innovation Research Complex Development Project of Busan Metropolitan City (IURP2401). Vision inspection refers to a series of inspections involving the acquisition of necessary images using imaging devices such as cameras, and the interpretation and judgment of the acquired images by mechanical devices such as computers in place of humans, as well as the execution of various measures. Such inspection devices are widely used for quality inspection of mass-produced products. Visual inspection is a type of quality inspection for mass-produced products. Products are inspected by optically analyzing images of their appearance to identify suspected defects and classify the defective products. For example, methods and devices that photograph the surface and extract images of the defects from the captured images are widely used to detect defects on the surface of a product under inspection. In particular, the use of pre-trained artificial intelligence models in this visual inspection process has improved inspection performance, reaching a level where reliable inspection results can be obtained even without human visual inspection. Specifically, the adoption of such machine vision-based visual inspection is increasing efficiency in terms of time and cost for inspecting mass-produced products. A practical issue in manufacturing sites is that while product surface inspection is a critical aspect for ultra-precision machining companies, small and medium-sized enterprises face labor shortages due to a lack of inspection personnel. Results can vary depending on factors such as employee turnover, differences in subjective judgment, eye strain, and the inspector's physical condition. However, there are currently products for which it is difficult to employ visual inspection methods with vision inspection systems. Products with glossy surfaces, such as automobiles and related parts requiring a glossy coating, and home appliances, have been difficult to inspect using machine vision because they reflect the light essential for exterior imaging. There was a problem in that it was difficult to identify defects in areas within the product images where there was excessive reflection due to surface gloss, even if defects were present. Consequently, these products presented an inefficiency where defects had to be detected by human visual inspection. Therefore, it is necessary to develop a vision inspection device that can perform objective quality inspection, automatically collect data, drastically reduce inspection time, aggregate the status of inspection results in real time, and immediately collect products with surface defects on-site. FIG. 1 is a block diagram of a vision inspection device for detecting surface defects according to an embodiment of the present invention. Figure 2 is a diagram of the CNN structure. FIG. 3 is a schematic diagram illustrating the placement of a vision inspection device for detecting surface defects according to an embodiment of the present invention on a production line. FIG. 4 is a schematic diagram illustrating the operation of a vision inspection device according to an embodiment of the present invention. FIG. 5 is a block diagram schematically showing an image capturing device processing an image of a product to be inspected. FIG. 6 is a flowchart illustrating a vision inspection method for detecting surface defects according to an embodiment of the present invention. Figure 7 is a flowchart illustrating a vision inspection method using light and vision AI. FIG. 8 is a flowchart for explaining the light projection method in more detail. FIG. 9 is a block diagram illustrating a computing environment including a computing device suitable for use in exemplary embodiments. Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. Before describing the present invention, it should be noted that in assigning reference numerals to the components of each drawing, the same components are given the same reference numeral whenever possible, even if they are shown in different drawings. In addition, in describing the present invention below, if it is determined that a detailed description of related known functions or configurations could unnecessarily obscure the essence of the present invention, such detailed description is omitted. Furthermore, it should be noted in advance that the terms used in this a