CN115372380A - Identification system for extremely-short wave optical detection method of plastic film wrapped outside - Google Patents

Abstract

The invention belongs to the field of film detection, and particularly relates to an identification system of an outsourcing plastic film ultra-short wave optical detection method. The identification system comprises an extremely short wave light source module, an image acquisition module, an image filtering preprocessing module and an image algorithm processing module, is used for detecting the defects of the externally-coated plastic film and transmitting the detection result to a result output module. The light source outline is detected by using the ultra-short wave light source, the light source grouping module for accurately measuring and calculating time is established, so that light rays are prevented from being refracted and reflected, the mutual interference of the light sources is prevented, and meanwhile, the quality of a product oil seal with a plastic film is better detected by using a film recognition algorithm and detecting and tracking creases.

Description

An identification system of ultra-short-wave light detection method for outsourcing plastic film

technical field

The invention belongs to the field of film detection, and in particular relates to an identification system for outsourcing plastic films with an ultrashort-wave light detection method.

Background technique

At this stage, plastic film has been widely used in food, medicine, chemical industry and other fields, among which food packaging accounts for the largest proportion, such as beverage packaging, quick-frozen food packaging, cooking food packaging, fast food packaging, etc. These products are given to people. Life brings great convenience.

However, in the appearance inspection based on the film, it is impossible to accurately judge the flaws of the product close to the human eye, and the machine vision system inspection has emerged as the times require. , Damage, edge cracks, wrinkles, dark spots, bright spots, edge damage, black spots, hairs, mosquitoes, and other common surface defects are detected and analyzed, and defective products can be quickly screened out, which greatly saves manpower and material resources.

However, in the actual production process of outsourcing plastic film, due to the influence of various factors, defects such as holes, mosquitoes, black spots, crystal spots, scratches, spots, etc. will appear on the surface of the film, which seriously affects the quality of the film. Unnecessary loss has been brought about. The traditional outer packaging plastic transparent film will refract and reflect light during the light irradiation process, and interference will occur during the film contour detection process, which will affect the accuracy of film contour detection and defect identification. There is a certain impact on the accuracy, so it is urgent to introduce a visual inspection method to better prevent the mutual interference of light sources, and at the same time, it can accurately measure the time when the light sources are grouped and turned on successively, so as to better detect the defects existing in the transparent plastic film of the outer packaging .

Contents of the invention

Aiming at some problems existing in the prior art, the present invention provides an identification system for ultra-short-wave light detection method of outsourcing plastic film, which specifically includes an ultra-short-wave light source module, used to control the light source switch and the light source bright and dark time, and through the image acquisition module Collect film profile data, and transmit the collected film profile data to the image filtering preprocessing module for filtering and distortion correction processing on the collected film profile data, and transmit the processed data to the image algorithm processing module for use in To detect the defects of the outsourcing plastic film, and transmit the detection results to the result output module.

Preferably, an infrared sensor is used to locate the position of the product with the plastic film, so that when the product with the plastic film is carried to a designated position through the assembly line, the light source camera is triggered to take pictures and identify it.

Preferably, the ultra-short-wave light source module includes a light source time controller and an ultra-short-wave light source.

Preferably, the light source time controller measures and calculates the light source time, and groups them to determine their order of brightness and darkness, so as to deal with the refraction, reflection and mutual interference after the ultra-short-wave light irradiates the plastic outer packaging film.

Preferably, the image filtering preprocessing module includes image acquisition, brightness contrast preprocessing, distortion correction, and rotation transformation processing.

Preferably, the detection of defects in the outer plastic film includes dense wrinkle detection, crease detection and tracking, damage detection, unreasonable shape and style screening of creases, detection of side film beyond the edge edge level difference detection, edge crease bend of transparent film image Refraction detection.

Preferably, the detection of the bending degree of the edge crease of the transparent film image uses RTU capture measurement technology.

Preferably, the image algorithm processing module includes film recognition algorithm, edge feature extraction, and shape matching based on deep learning.

Preferably, the edge feature extraction firstly uses canny operator to extract preliminary edge features, and then performs Laplacian operation, sobel operation and MID-EDGE operation to extract edge depth features.

Preferably, the canny operator extracts preliminary edge features, uses denoising to filter areas where the gray value of the edge of the outer film changes sharply, and then derivates the obtained gradient value to represent the gray value of the edge of the outer film The degree of change, and through non-maximum value suppression and threshold value processing, it is used to fit the edge line and optimize the edge characteristics of the outsourcing film.

Compared with the prior art, the present invention has the following beneficial effects:

(1) The present invention uses a very short-wave light source to detect the light source profile, and establishes a light source grouping module that accurately measures the time to avoid light refraction and reflection and prevent mutual interference of light sources.

(2) The present invention uses a film recognition algorithm to extract the deep features of the film edge by detecting and tracking creases, damage detection and unreasonable crease shape collection screening, etc., so as to better detect the oil seal of products with plastic films quality.

In the process of detecting the outline of the outer packaging film, by using a very short-wave light source

Description of drawings

Fig. 1 is a detection flow chart of the ultra-short-wave light detection method of the outsourcing plastic film.

Detailed ways

The present invention provides a multi-dimensional data acquisition and detection method combined with real sensory simulation. In order to make the purpose, technical solution and effect of the present invention clearer and clearer, the present invention will be further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.

An identification system for the ultra-short-wave light detection method of outsourcing plastic films, which specifically includes an ultra-short-wave light source module, which is used to control the light source switch and the light source's bright and dark time, and collects film profile data through the image acquisition module, and transmits the collected film profile data to To the image filtering preprocessing module for filtering and distortion correction processing on the collected film profile data, and sending the processed data to the image algorithm processing module to detect the defects of the outer plastic film, and sending the detection results to the results output module.

In one embodiment, an infrared sensor is used to locate the position of the product with the plastic film, so that when the product with the plastic film is carried to a designated position through the assembly line, the light source camera is triggered to take pictures and identify it.

In one embodiment, the ultra-short-wave light source module includes a light source time controller and an ultra-short-wave light source.

In one embodiment, the light source time controller calculates the light source time and groups them to determine the order of their brightness and darkness, so as to deal with the refraction, reflection and mutual interference after the ultra-short-wave light irradiates the plastic outer packaging film.

In one embodiment, the image filtering and preprocessing module includes image acquisition, brightness contrast preprocessing, distortion correction, and rotation transformation processing.

In one embodiment, the detection of the defects of the outer plastic film includes wrinkle-intensive detection, crease detection tracking, damage detection, unreasonable shape and style screening of creases, side film beyond the edge edge detection, transparent film image Edge crease bending detection.

In one embodiment, the detection of the bending degree of the edge crease of the transparent film image uses RTU capture measurement technology.

In one embodiment, the image algorithm processing module includes thin film recognition algorithm, edge feature extraction, and shape matching based on deep learning.

In one embodiment, the edge feature extraction first uses a canny operator to extract preliminary edge features, and then performs Laplacian operations, sobel operations, and MID-EDGE operations to extract edge depth features.

In one embodiment, the canny operator extracts preliminary edge features, and uses denoising to filter out areas where the gray value of the edge of the outer film changes sharply, and then uses the derived gradient value to represent the outer film The degree of change of the edge gray value is used to fit the edge line and optimize the edge characteristics of the outsourcing film through non-maximum value suppression and threshold value processing.

The above description is only a preferred embodiment of the present invention, and is not intended to limit the invention in other forms. Any skilled person who is familiar with this field may use the technical content disclosed above to change or change into an equivalent embodiment with equivalent changes. However, any simple modifications made to the above embodiments according to the technical essence of the present invention, equivalent changes and modifications, still belong to the protection scope of the technical solutions of the present invention, as long as they do not depart from the content of the technical solutions of the present invention.

Claims (10)

1. An identification system of ultra-short-wave light detection method for outsourcing plastic films, specifically including an ultra-short-wave light source module, which is used to control the light source switch and the light source’s bright and dark time, and collect film profile data through the image acquisition module, and collect the collected film profile data , sent to the image filtering preprocessing module for filtering and distortion correction processing on the collected film profile data, and sending the processed data to the image algorithm processing module for detecting the defects of the outer plastic film, and sending the detection results to the result output module. 2. The identification system of an outsourced plastic film ultra-short-wave light detection method according to claim 1, characterized in that an infrared sensor is used to locate the position of the product with the plastic film, so that when the product with the plastic film is carried to the specified location through the assembly line After the location, trigger the light source camera to take pictures and identify them. 3 . The identification system of an outer-wrapped plastic film using an ultra-short-wave light detection method according to claim 1 , wherein the ultra-short-wave light source module includes a light source time controller and an ultra-short-wave light source. 4. The identification system of an outsourced plastic film ultra-short-wave light detection method according to claim 3, wherein the light source time controller determines the order of light and dark by measuring and calculating the light source time, and grouping them, for Dealing with the refraction, reflection and mutual interference after the extremely short-wave light irradiates the plastic outer packaging film. 5. The identification system of a kind of outsourcing plastic film ultra-short-wave light detection method according to claim 1, characterized in that, the image filtering preprocessing module includes image acquisition, brightness contrast preprocessing, distortion correction, and rotation transformation processing . 6. The identification system of an outer-wrapped plastic film with an ultra-short-wave light detection method according to claim 1, wherein the detection of defects in the outer-wrapped plastic film includes wrinkle-intensive detection, crease detection and tracking, damage detection, and fold detection. Screening of unreasonable shape and style of marks, detection of side film beyond the edge position difference, and detection of edge crease bending degree of transparent film image. 7 . The identification system of an outer-wrapped plastic film with an ultra-short-wave light detection method according to claim 6 , wherein the detection of the edge crease and bending degree of the transparent film image uses RTU capture measurement technology. 8. The identification system of an outer-wrapped plastic film with an ultra-short-wave light detection method according to claim 1, wherein the image algorithm processing module includes a film identification algorithm, edge feature extraction, and shape matching based on deep learning. 9. A kind of outsourcing plastic film ultra-short-wave light detection method recognition system according to claim 8, is characterized in that, described edge feature extraction, first uses canny operator to extract preliminary edge feature, and carries out Laplacian operation , sobel operation and MID-EDGE operation to extract edge depth features. 10. The identification system of a kind of outsourcing plastic film extremely short-wave light detection method according to claim 1, characterized in that the canny operator extracts preliminary edge features, and is used to filter out the sharp gray value of the edge of the outsourcing film through denoising In the changed area, the obtained gradient value is derived to represent the change degree of the gray value of the edge of the outer film, and through non-maximum value suppression and threshold value processing, it is used to fit the edge line and optimize the edge of the outer film feature.

Images