Visual all-angle imaging device in high-reflection mirror surface environment
Technical Field
The invention relates to the technical field of optical imaging, in particular to a visual all-angle imaging device in a high-reflection mirror surface environment.
Background
The AI appearance vision inspection technology is through high efficiency, high fitness computer vision technology-deep learning, reinforcement learning, algorithm components, knowledge bases, etc., in cooperation with optical imaging schemes. The high-efficiency and high-adaptability computer vision technology-deep learning, self-building self-adaption, high-efficiency acceleration and flexible expansion infrastructure system has the advantages that the front end bears multi-source data, the rear end supports industrial application, intelligent collection, cleaning, processing, management, analysis, operation and maintenance, monitoring and visualization are carried out on the data, and therefore intelligent transformation of the manufacturing industry is enabled at higher speed. The AI appearance visual inspection technique comprises two major cores: AI algorithm, optical imaging technique.
At present, in a traditional optical imaging scheme, when the defect detection is performed on a high-reflection mirror surface material, due to the optical physical characteristics of the material, the phenomena of multi-angle reflection and scattering (almost all angles in a 90-degree included angle range formed by a light source incidence angle and a camera view angle) exist, a camera and a light source are erected at a single angle, and due to the angle limitation, only a part of defect characteristic images can be captured, so that the camera and the light source can be erected at multiple angles. However, due to the mutual interference of multiple light sources, the limitation of hardware size, production space and hardware cost, the traditional optical imaging scheme cannot capture all reflected light and scattered light on the surface of the material all the time, and cannot finish the defect full-feature presentation of the high-reflectivity mirror surface material.
The existing optical imaging scheme adopts a camera and a light source which are erected at multiple angles when detecting the defects of a high-reflector surface material. However, due to the mutual interference of multiple light sources, the limited size of hardware, limited erection space, limited hardware cost and the like, all reflected light and scattered light on the surface of the material cannot be captured all the time, and the defect full-characteristic presentation of the high-reflector surface material cannot be completed, so that the vision full-angle imaging device in the high-reflector surface environment is provided for this reason
Disclosure of Invention
Technical problem to be solved
Aiming at the defects of the prior art, the invention provides a visual full-angle imaging device in a high-reflection mirror surface environment, which is mainly used for detecting and imaging surface defects of a high-reflection mirror surface material, the defect detection of the material still depends on manual work seriously in an actual application scene at present, an operator needs to change angles continuously under a strong light environment to search for different reflection characteristic angles, and the problem that in the prior art, due to the space and cost limitation of an actual application scene, the optical coverage angle of the manual detection cannot be realized, the light escape phenomenon is caused, and part of defects cannot be imaged successfully is solved.
(II) technical scheme
The invention provides the following technical scheme: the visual full-angle imaging device comprises a macro camera array, wherein the macro camera array comprises a first camera array, a second camera array and a third camera array, the macro camera array is fixedly installed below an image data line channel support, lateral light sources are arranged on the left side and the right side of the macro camera array, a product carrier is arranged in the area below the image data line channel support, and the product carrier moves back and forth below the macro camera array through a translation module.
Furthermore, the shooting area of the second camera array is vertically downward, the shooting angle areas of the first camera array and the third camera array are inclined towards the middle, the three groups of camera arrays are arranged, the number of the arrays is adjusted according to the size of the detected material, the visual fields between the cameras are mutually covered, the cameras are close enough to the shot material, and the reflected light and the scattered light of any angle can escape from the imaging range of a single camera but still be captured by the adjacent cameras.
Furthermore, the lateral light source comprises a first lateral light source and a second lateral light source, the first lateral light source and the second lateral light source are identical in specification, the lateral light source uniformly irradiates the shooting area, and the brightness of the shooting area is improved through the arrangement of the lateral light sources, so that the imaging effect of the device is better.
Further, the bottom of macro camera array is fixed and is equipped with the bottom support, the both sides at the top of bottom support are all fixed and are equipped with L shape frame, first lateral light source and second lateral light source are installed respectively on two L shape frames, and through the setting of L shape frame, be convenient for install respectively fixedly to first lateral light source and second lateral light source from both sides.
Furthermore, the translation module is including translation slider and translation track, the product carrier passes through the translation slider and slides on the translation track, and when the product carrier passed through the translation module and removes, the translation slider can slide on the translation track, through translation slider and the orbital matching of translation with realize the removal of translation module.
Further, the use method of the device comprises the following steps:
and S1, transferring and positioning the detected products from the front end conveying channel by a robot or manually, wherein the product carrier is used for installing and placing the products to be detected, and the products to be detected are placed on the product carrier and then move to the lower part of the camera array along with the product carrier for detection.
S2, the product carrier is driven by the translation module to move towards the camera array, and the translation module is convenient for moving the product carrier to the lower part of the camera array for detection.
And S3, when the product carrier bears the detected products and reaches the edge of the shooting range of the camera array, the translation module is changed from constant motion to lattice-by-lattice start-stop forward motion.
S4, uniformly irradiating the shooting area by the 'side light source', starting the 'camera array', and shooting one by one.
And S5, acquiring the image captured by the camera array through an AI algorithm system, identifying the defect, and labeling the corresponding position on the image spliced by the camera array.
And S6, when the product carrier bears the detected products and is separated from the shooting area, changing the translation module from the step-by-step start-stop forward motion into the uniform motion, conveying the detected products forward, and removing the detected products from the product carrier by a robot hand or a manual work.
S7, the product carrier carried by the translation module is reset to wait for the position of the subsequent detected product.
Compared with the prior art, the invention provides a visual all-angle imaging device in a high-reflection mirror surface environment, which has the following beneficial effects:
1. the invention relates to a visual full-angle imaging device in a high-reflection mirror surface environment, which uses a micro-distance camera array (the number of the array is adjusted according to the size of a detected material) and a single light source for imaging, the visual fields between cameras are mutually covered, the distance between the cameras and the shot material is close enough, and the reflected light and the scattered light at any angle can still be captured by adjacent cameras although the reflected light and the scattered light can escape from the imaging range of the single camera. Therefore, the full-characteristic presentation (imaging) of the surface defects of the high-reflectivity material is realized, and the problem that the surface defects of the material cannot be presented completely by optical hardware due to the light escape phenomenon in the conventional optical scheme is thoroughly solved.
2. According to the visual all-angle imaging device in the high-reflection mirror surface environment, the volume of the used macro camera is only 10% -15% of that of the currently popular industrial camera, so that an array covering the visual field range mutually is easy to form, the distance between the macro camera and a shot object is close enough (15-20mm), and the light escape is physically blocked; if the prior art is used to theoretically realize the view coverage, the volume and the cost are 10 times or more than the invention, and due to the limitations of the shooting distance (100 mm and 300mm) and the camera volume, and the space limitation of practical production, the requirement of high speed, the full coverage of the angle can be realized only in a laboratory environment.
Drawings
FIG. 1 is a three-dimensional view of the structure of the present invention;
FIG. 2 is a side view of the structure of the present invention;
FIG. 3 is a top view of the structure of the present invention.
Wherein: 1. a first camera array; 2. a second camera array; 3. a third camera array; 4. an image data line channel support; 5. a first lateral light source; 6. a second lateral light source; 7. a product carrier; 8. a translation module.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1-3, a visual full-angle imaging device of a high-reflectivity mirror environment comprises a macro camera array, the macro camera array comprises a first camera array 1, a second camera array 2 and a third camera array 3, the shooting area of the second camera array 2 is vertically downward, the shooting angle areas of the first camera array 1 and the third camera array 3 are inclined towards the middle, the volume of the macro cameras used in the invention is only 10% -15% of that of the currently popular industrial cameras, so that the micro camera array is easy to form an array covering the visual field range mutually, the distance between the micro camera array and the shot object is close enough to 15-20mm, and the occurrence of light is blocked physically, although the reflected light and the scattered light at any angle can escape from the imaging range of a single camera, the reflected light and the scattered light can still be captured by adjacent cameras, so that the full-characteristic imaging of the surface defect of the high-reflectivity material is realized, the problem that the surface defect of the material cannot be completely presented by optical hardware due to the light escape phenomenon in the existing optical scheme is thoroughly solved, a macro camera array is fixedly arranged below an image data line channel support 4, lateral light sources are arranged on the left side and the right side of the macro camera array, each lateral light source comprises a first lateral light source 5 and a second lateral light source 6, the specifications of the first lateral light source 5 and the second lateral light source 6 are the same, the lateral light sources uniformly irradiate a shooting area, a bottom support is fixedly arranged at the bottom of the macro camera array, L-shaped frames are fixedly arranged on the two sides of the top of the bottom support, the first lateral light source 5 and the second lateral light source 6 are respectively arranged on the two L-shaped frames, the lower area of image data line channel support 4 is equipped with product carrier 7, product carrier 7 passes through translation module 8 and moves in the below back-and-forth movement of macro camera array, translation module 8 is including translation slider and translation track, product carrier 7 slides on the translation track through the translation slider, when product carrier 7 moves through translation module 8, the translation slider slides on the translation track, through translation slider and the orbital matching of translation with realize the removal of translation module 8.
S1, conveying the detected product from the front end conveying channel, wherein the detected product is a high-reflection mirror surface material, and the detected product is moved and positioned on the product carrier 7 by a robot or a manual work, so that the detected product can be placed on the surface of the fixed carrier 7 and then moves to the lower part of the camera array along with the product carrier 7 for detection.
S2, the product carrier 7 is driven by the translation module 8 to move in the direction of the camera array, the product carrier 7 is convenient to move to the lower part of the camera array under the action of the translation module 8, and then the product to be detected on the product carrier 7 is convenient to shoot and detect through the camera array.
S3, when the product carrier 7 bears the detected product and reaches the edge of the shooting range of the camera array, the translation module 8 changes the uniform motion into the lattice-by-lattice start-stop forward motion.
S4, uniformly irradiating the shooting area by the 'side light source', starting the 'camera array', and shooting one by one.
And S5, acquiring the image captured by the camera array through an AI algorithm system, identifying the defect, and labeling the corresponding position on the image spliced by the camera array.
And S6, when the product carrier 7 bears the detected products and is separated from the shooting area, changing the translation module 8 from the step-by-step start-stop forward motion into the uniform motion, conveying the detected products forward, and removing the detected products from the product carrier 7 by a robot hand or a manual work.
S7, the product carrier 7 carried by the translation module 8 is reset to wait for the next detected product to be in place.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.