• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    System and method of detection of defects on specular or semi-specular surfaces by photogrammetric projection. The present invention relates to a system for detecting defects on specular or semi-specular surfaces of objects to be inspected, suitable for integration into an industrial production line, comprising: a plurality of light emission means (1) on the objects; a plurality of cameras (2) for detecting the light reflected by the objects; a photogrammetric control and analysis subsystem of information associated with the light emitted by the light emission means (1) and the light detected by the cameras (2); and a support structure (3), integrated in the production line of the objects to be inspected, in which the light emission means (1) and the cameras (2) of the system are arranged. Advantageously, the light emission means (1) are arranged in the system in such a way that the light emitted on the surface of the objects has a periodic pattern of light and shadow, and said pattern is emitted with a relative movement on the objects a inspect, covering its surface. (Machine-translation by Google Translate, not legally binding)
    公开号:ES2630736A1
    申请号:ES201531776
    申请日:2015-12-07
    公开日:2017-08-23
    发明作者:Juan José Aguilar Martín；Jorge Santolaria Mazo；David SAMPER CARNICER；Jesús VELÁZQUEZ SANCHO
    申请人:Universidad de Zaragoza；
    IPC主号:
    专利说明:

    5
    10
    fifteen
    twenty
    25
    30
    35
    DESCRIPTION
    SYSTEM AND METHOD OF DETECTION OF DEFECTS IN SPECULAR OR SEMI-SPECULAR SURFACES THROUGH PHOTOGRAPHIC PROJECTION
    FIELD OF THE INVENTION
    The present invention falls within the technical field corresponding to the analysis of surfaces for the detection and localization of defects or heterogeneities. Its scope of application refers mainly to industrial processes where the final products require a paint finish, as is the case in the automotive, transportation, appliances, etc. More specifically, the invention relates to a system for detecting and locating defects in specular or semi-specular surfaces of objects, by means of photogrammetric projection, which can be used in real time in an industrial manufacturing or assembly line. The invention also relates to a method of detecting defects based on said system.
    BACKGROUND OF THE INVENTION
    Paint treatments are commonly used to improve the surface characteristics of industrial products, such as their resistance to corrosion, wear, improve the adhesion of other products, tightness to prevent water ingress, and aesthetic properties such as color or the bright.
    Once the paint treatment of these products is finished, they are usually inspected within the framework of the quality control processes. This inspection, in many cases, is usually visual and performed by operators in adequate facilities, to have the required and necessary lighting. However, the final validation decision always depends on the vision and analysis of the person responsible for the inspection, therefore there are many conditions or uncontrolled situations (fatigue, mood or distractions) that can lead to some defects they are not detected correctly, remaining in the final product that reaches the customer.
    Once a product with a defective finish is already outside the production line, the repair of the surface finish entails greater complexity
    5
    10
    fifteen
    twenty
    25
    30
    35
    Logical and operational, with the consequent impact on production costs. Therefore, if defect detection is carried out efficiently in a first control over the production line, the associated cost will be much lower.
    The defects to be identified can generally be of two types: those that produce local variation of the surface geometry and those that produce a local variation of contrast or color.
    The defects of the first group are spots or specks that have been adhered to the product before the painting process, being the most common defect. It is produced by the inclusions of dirt, foreign bodies and dust, which cause small and granular inequalities, almost always existing in large quantities and distributed more or less regularly, modifying the surface geometry locally. These defects are best detected where they become more visible when observed, when they are illuminated with structured light, near a border of light-dark light. That change in the light that is reflected in that area is what makes it possible, using artificial vision techniques on the images taken by cameras, to detect the defect. Structured light reflection is essentially sensitive to curvature and, therefore, allows the detection of geometry variation defects that are imperceptible using traditional artificial vision techniques or triangulation techniques with direct projection on surfaces. The principle of operation is based on the generation of relative movement between the surface to be examined and the light with which they are illuminated.
    The defects of the second group are small scratches, spots, etc., which can also be detected by diffuse illumination and gradient or contrast algorithms.
    There are systems for detecting dents or undulations, of a typical depth of a few microns, intended for the inspection of vehicle body plates, based on the use of 3D optical sensors and signal processing means, as described in the article "Visualization and Detection of Small Defects on Car-Bodies" (S. Karbacher et al.), published in Vision, Modeling and Visualization '99 (Proceedings), Infix, 18 (1999). These systems can comprise a portico integrated in a car production line, where a plurality of light emission sources and cameras are installed to detect the light reflected by the surfaces of the cars. They also incorporate a means of acquiring the data generated by the emission sources and by
    5
    10
    fifteen
    twenty
    25
    30
    35
    the cameras, as well as a process subsystem and analysis of surface defects of the cars to be inspected.
    On the other hand, there are also references to real-time photogrammetry systems, based on short-range digital photogrammetry, for the analysis of deformations in automobile bodies in the assembly line, such as the one described in the article "Automated Dimensional Inspection with Realtime Photogrammetry ”(HA Beyer), published in ISPRS Journal of Photogrammetry and Remote Sensing, Volume 50, Issue 3, p.20-26 (1995) .This system is based on the use of CCD cameras for car inspection, together with hardware and software used to analyze the data obtained.
    Although the systems described above and their associated procedures serve to detect surface heterogeneities in the products by different techniques, the main problem of all of them is that they comprise a high error rate in the identification of defects of small dimensions, due to the interference generated by vibrations and noise due to movement in the production line. Likewise, the aforementioned systems are not suitable for the analysis of products of mirror or semi-specular surfaces of different colors in the same production line.
    On the other hand, a high contrast structured light system is necessary, capable of enhancing the geometric defect by an optical amplification phenomenon in the vicinity of the border of the illumination, and which allows defects to be detected with greater precision and greater resolution.
    The need arises, therefore, to solve the problems currently existing in the known detection procedures, to improve the accuracy and quality of the image obtained and thus obtain more precise results in the detection of defects, in turn reducing the appearance of false positives.
    BRIEF DESCRIPTION OF THE INVENTION
    An object of the present invention is, therefore, the obtaining of defect detection systems that allow obtaining accurate results, improving defect detection rates on the objects to be inspected. For this, the invention proposes the use of redundancy patterns in the gradients of the structured light used, and an adequate capture rate of the cameras, so that two-dimensional (2D) filters can be used and
    5
    10
    fifteen
    twenty
    25
    30
    35
    three-dimensional (3D), by means of photogrammetric projection, to eliminate the vibrations, reflections and noises present in the production line of the objects. In the factory inspection, normally the object to be inspected is in motion, due to the displacement of the production line itself, the light patterns remaining static. This allows the relative movement between product and light to explore the entire surface of the object which, together with the appropriate treatment, detection and tracking algorithms, allows inspection without interfering with the normal movement of the products in the manufacturing line. .
    Said object of the invention is preferably obtained by means of a system for detecting defects in specular or semi-specular surfaces of objects to be inspected, suitable for integration into an industrial production line, comprising:
    - a plurality of light emitting means on the objects;
    - a plurality of cameras for the detection of the light reflected by the objects;
    - a photogrametric control and analysis subsystem of information associated with the light emitted by the light emitting means and the light detected by the cameras; Y
    - a support structure in which the light emitting means and the cameras of the system are arranged.
    Advantageously, and to generate the redundant analysis of the defects, the light emission means are arranged in the system so that the light emitted on the surface of the objects has a periodic pattern of light and shadow, where said pattern is emitted with a relative movement on the objects to inspect, crossing its surface. The relative movement existing between the periodic pattern of light / shadow and the objects to be inspected is produced, either by the movement of said objects, the static emitted pattern remaining on them, or by the movement of the pattern emitted on the objects to be inspected. , while said objects remain static.
    In a preferred embodiment of the invention, the cameras are capture cameras with an image acquisition frequency between 40-100 Hz. More preferably, the cameras are industrial cameras of the CCD or CMOS type.
    In another preferred embodiment of the invention, the light emitting means are means of direct illumination on the objects to be inspected. More preferably, said means comprise neon light emitters, fluorescents, or light emitting diodes (LEDs).
    5
    10
    fifteen
    twenty
    25
    30
    35
    In another preferred embodiment of the invention, the light emitting means are means of indirect illumination on the objects, by means of projection or rear projection on one or more screens. More preferably, the indirect illumination means emit periodic patterns of light and shadow on the screens which, in turn, are reflected on the objects to be inspected.
    In another preferred embodiment of the invention, the information control and analysis subsystem comprises process and treatment means by photogrammetric projection of the images captured by the cameras, 2D and / or 3D filtering means for the identification of defects in said images, means of analyzing the color of the objects to be inspected, and / or means of representing the objects and defects identified by augmented reality.
    In another preferred embodiment of the invention, the system also comprises one or more of the following elements: high resolution lenses for cameras, one or more computers for processing and control of the system, one or more measurement elements to control position of the objects and synchronize the catches of the cameras.
    Another object of the invention relates to a method of identifying defects on surfaces of objects to be inspected, which comprises the use of a system according to any of the embodiments described herein, where said system is optionally integrated into a production line. industrial objects to inspect.
    The method of the invention comprises performing at least the following steps:
    - the surface of the objects to be inspected is illuminated with the light-emitting means, so that the light emitted on said surface has a periodic pattern of light and shadow, with a relative movement on the objects to be inspected;
    - the light reflected by the objects is registered by means of the cameras;
    - the reflected light and the light emitted through the control and analysis subsystem are analyzed, through photogrammetric projection for the detection of surface defects.
    In a preferred embodiment of the method of the invention, the periodic pattern of light and shadow has an adaptive shape to the surface of the objects to be inspected, so that said pattern varies depending on the curvature of the inspected surface.
    5
    10
    fifteen
    twenty
    25
    30
    35
    In general, the system and method for the detection of defects of the invention solves the technical problem of achieving the localization and identification of the type of defect in the production line itself, by means of a robust and optimal system, optionally incorporating tools of augmented reality for subsequent efficient repair by an operator.
    The system includes a series of cameras located so that they cover the entire surface to be investigated, where said cameras are prepared for the acquisition of high-speed images, which will then be treated using artificial vision algorithms in real time.
    As mentioned, the lighting means consist of a set of light beams, or structured, high contrast patterns of light capable of enhancing the geometric defect by an optical amplification phenomenon in the vicinity of the border of the lighting, it is that is, in the transits from light to dark and from dark to light newspapers of the lighting pattern used. In these transits, the reflection of the light varies due to the local variation of curvature, which allows to highlight the defects with a diameter of up to ten times the size of the defect itself. It is this phenomenon that enables the system to detect small defects of micrometric size. It is necessary that the light that hits the surface be of high contrast, for example by light / shadow patterns. Said set of light beams can be created by high frequency fluorescent tubes, LED illumination, or by projection or retro-projection of images with high contrast patterns on a screen that is reflected on the surface.
    Illumination is one of the most important components of the system, since the success of a good recognition of the defect in the image depends largely on a good reflection of the light on the surface. As described in previous paragraphs, the invention raises two possible forms of illumination: direct and indirect.
    The first lighting system, direct lighting, is carried out through the use of neon tubes or high frequency LED or fluorescent lighting, fixed to the structure of the system (inspection tunnel, for example). The reflection of this light generates a redundant pattern of fixed lines on the mobile surface of the object, allowing a complete sweep of the surface of the objects.
    5
    10
    fifteen
    twenty
    25
    30
    35
    The second system, indirect lighting, works by means of projectors. The basic principle is to project or retro-project different structured high contrast light patterns on a screen located in the vicinity of the object to be inspected and observe the scan of the patterns reflected in the mobile or fixed object. With this technique, different light patterns can be generated, such as points, straight lines, circumferences, crosses, squares, triangles, etc., being able to select the most appropriate lighting depending on the shape of the object to be controlled. The light patterns may be fixed, adaptive to the surface, moving to certain positions to sweep or sequentially a light-dark pattern of the forms described followed by their dark-light inverse pattern. As previously mentioned, adaptive patterns vary the size of the lattice as a function of the curvature of the surface to be inspected, so that the areas near the light-dark borders are of the smallest size suitable for localization of the defects. The design of these patterns can also be done in a way that maximizes the number of amplifications per number of images so that the surface is scanned with the minimum number of images. Both the shape and size of the pattern grid can be changed throughout the inspection to adapt to the curvature of the surface, when the product is in motion, or to sweep the entire surface when the product remains static.
    The cameras are the other fundamental part of the system. These are arranged so that the total coverage of the area to be inspected is possible by inspection without contact, without interfering with the normal movement of the products through the manufacturing assembly line, and so that the light reflection angles maximize the detection of surface defects.
    The defect detection system remains static on the support structure of the system. The products to be inspected are those that can carry the movement (normal movement of the line without interfering with the speed of production and without the need for desvms of the products) for scanning. These displacements of the products are obtained by means of a measurement system that captures the movement of the production line. Therefore, the system can be incorporated into the production line so that the time of the production process is not increased by the inspection of the products.
    Another alternative proposed by the invention is to inspect static products, in which case it is necessary for the lighting means to sweep or cross the entire surface to be inspected.
    5
    10
    fifteen
    twenty
    25
    30
    35
    For the capture of the images, industrial vision cameras are available, whose high capture frequency makes it possible to detect defects even if there are possible vibrations associated with the movement of the products in their inspection. To determine the number and type of cameras that are part of the system, the maximum speed of the production line, the maximum width to be inspected, the different changes or inclinations from the normal to the surface and the size of defect must be taken into account. mmimo to detect.
    The system is configured to capture images in a synchronized way, between the cameras and the movement of the inspected product, as it travels in the production line. Before possible stops of the production line, the system has a measuring element (encoder, distance meter, speedometer, vision camera, etc.) to control the position of the object and synchronize the capture of the cameras, so that it is not loses synchronism at any time of inspection.
    In order to optimally capture the images of products of different colors, the possibility of adapting the capture parameters (gain, exposure time, etc.) to the specific color of the inspected product is incorporated into the system, keeping those capture parameters updated in the time and therefore reactive to possible variations in the sensitivity of the cameras and in the luminance of the lighting systems.
    After image captures, these are processed in real time by optimized algorithms for image processing (image smoothing, highlighting gray level gradients or contour detection) that allow defects to be detected by local level variations of gray
    Masks are also applied to treat only the areas of interest of the captured images, thus increasing the speed of the process and avoiding reflections of unwanted areas.
    With the objective that the system has a high capacity for locating defects and does not present false positives, during the 2D inspection the possible defect detected is monitored, taking into account both its physical behavior and its optical properties in consecutive images. In turn, the 2D spatial information obtained in this process is treated to obtain the 3D spatial approximation of the defect detection, for this purpose different groups of homogeneous and symmetric polynomials formed by monomials with different heterogeneous variables, both spatial and temporal, are used. Crossing this information and applying different treatments
    5
    10
    fifteen
    twenty
    25
    30
    35
    Statistics on the data obtained in both processes can be determined with sufficient confidence, if a positive detection of the system is due to a defect or a spurious reflection on the inspected surface, which allows filtering based on the 2D information eliminating false positives that may occur. In addition, obtaining the 3D coordinates of the defects by means of photogrammetric projection of the possible 2D defects on the nominal surface of the inspected product allows the improvement of the treatment, by integrating a 3D grouping filtering, with which it is possible to reinforce the performance of the system using high redundancy.
    The redundancy in the capture of the images is also used as a self-diagnostic procedure thanks to the overlaps of the fields of vision of the cameras. The defects detected in these overlap zones allow calculation of the mismatch errors of the cameras. When the difference between the 3D coordinates of the defects exceeds a threshold, the system can be configured to alert about the need to recalibrate the affected cameras.
    The analysis of the images is carried out by means of several data processing units and with a central control PC. The calculation mechanism of the processors performs all the arithmetic and logic operations with link to the data obtained. The control PC controls the system in such a way that all operations are performed in the order of temporal and logical succession.
    The main advantage of the invention is the large number of times a defect is detected thanks to the use of redundancy in the gradients of the structured light used and the capture rate of the cameras, so that the appropriate 2D and 2D filters can be used. 3D (by means of photogrammetric projection) so that the vibrations, reflections and noises present in the production line are eliminated. In the inspection the object normally carries the movement, displacement of the production line itself, the light patterns remain static. This allows the relative movement between product and light to scan or explore the entire surface of the object, which, together with the developed treatment, detection and tracking algorithms allows inspection without interfering with the normal movement of the products in the manufacturing line .
    Once the 3D coordinates of the defects, their cataloging and severity are obtained, the results are presented on the screen graphically or in augmented reality glasses, to facilitate the subsequent rapid repair of them.
    5
    10
    fifteen
    twenty
    25
    30
    35
    DESCRIPTION OF THE FIGURES
    Figure 1 shows a perspective view of the system of the invention, according to a preferred embodiment thereof based on direct illumination means, through fluorescent tubes.
    Figure 2 shows a perspective view of the system of the invention, according to a preferred embodiment thereof based on direct illumination means, through a tunnel of LEDs.
    Figure 3 shows a perspective view of the system of the invention, according to a preferred embodiment thereof based on indirect lighting means, through projection on screens.
    Figure 4 shows a perspective view of the system of the invention, according to a preferred embodiment thereof, where said system is integrated in a production line of the objects to be inspected.
    Figures 5-8 show lighting and projection patterns used in different embodiments of the present invention.
    DETAILED DESCRIPTION OF THE INVENTION
    Next, different examples of preferred embodiments of the invention are provided, provided for illustrative but not limiting purposes thereof. Referring to Figures 1-4 of this document, the surface defect detection system of the invention comprises:
    - a plurality of light emitting means (1) on the surfaces of one or more objects to be inspected;
    - a plurality of cameras (2) for recording the light reflected by said surfaces;
    - a static support structure (3), integrated in the production line of the objects to be inspected, in which the light emitting means (1) and the cameras (2) of the system are arranged;
    - a subsystem (not shown in Figures 1-4) for control and analysis of information associated with the light emitted by the light emitting means (1) and / or the light detected by the cameras (2).
    5
    10
    fifteen
    twenty
    25
    30
    35
    For the image capture of the surfaces to be inspected, a plurality of high frequency industrial vision cameras (2) of image capture per second, typically between 40-100 Hz, for example CCD or CMOS cameras, will preferably be used. The number and specific type of cameras that will be part of the system will depend on different parameters, such as the dimensions of the object to be inspected, maximum speed of the manufacturing chain and minimum size of the defect to be detected. For example, for the complete inspection of the bodywork of a vehicle, 25-30 CMOS 40 Hz cameras can be used to detect defects of up to 0.1 mm in a 130 mm / second speed manufacturing chain.
    As mentioned, the elements of the defect detection system will remain fixed on the support structure (3) (for example a portico structure in Figure 1), as the objects to be inspected move in the production line thereof. During this movement, the cameras (2) will acquire images of the objects, and then they will be processed by the control and analysis subsystem, obtaining three-dimensional information (3D) about the dimensions of the defects and their situation on the surfaces of the objects . For the taking of the images (that is, the adequate recording of the light reflected on the objects), it is necessary to use suitable light emission means (1), since the success of a good image recognition depends largely on Good lighting. Alternatively, the objects to be inspected may remain static, with the light-emitting means (2) being those that scan their surfaces.
    In a first preferred embodiment of the invention, the illumination is carried out by means of direct light emission (1) on the objects, that is, said light directly affecting the surfaces to be inspected. Said means may be, more preferably, neon tubes or high frequency fluorescent tubes (Figure 1), or means based on light emitting diodes (LEDs) (see Figure 2, where said technique configured as a lighting tunnel is used on the objects to inspect). An example of this type of lighting is shown in Figure 4, integrated into a system for detecting defects in automobile surfaces.
    In a second preferred embodiment of the invention (Figure 3), the light-emitting means (1) are means of indirect illumination on the objects, by means of projection or rear projection. The basic principle of this embodiment is to project or retro-project different light patterns on a screen (4) located in the vicinity of the object a
    5
    10
    fifteen
    twenty
    25
    30
    35
    inspect, and observe the scan of the patterns reflected on the surface of it. An example of the projection of a linear light pattern is shown in Figure 3, which behaves in an equivalent manner to the light provided by fluorescent tubes (Figure 1). Depending on the type of object to be inspected, the number of projectors that are necessary will be determined. For an example of inspection of automobile bodies, a minimum of 3 projectors is necessary (one for the analysis of horizontal surfaces and two for the lateral surfaces of the car, as shown in Figure 3, also with a support structure ( 3) portico type).
    The main advantage of the present invention is obtained by the provision of the light emission means (1), both in direct and indirect lighting, so that a pattern of periodic and high contrast lighting is generated on the surfaces to be inspected. This pattern generates periodic illumination with areas of light and shadow, which allows the control and analysis system the successive calculation of multiple detections in the light / shadow border regions, providing a redundant analysis that helps both increase the detection rate of defects, such as limiting the appearance of false positives.
    Depending on the shape and arrangement of the direct lighting means, or the selected projection pattern, it is possible to generate different contrast patterns (points, lines, circumferences, crosses, squares, etc.) depending on the geometric particularities of the object to inspect. In the event that the inspected product is in motion in the manufacturing line, the patterns may change in size and adapt to the curvature of the inspected surface, so that the areas near the contrast borders have the necessary minimum size in the images to detect the defects. Likewise, in the event that the product remains immobile during its inspection in the manufacturing line, the light patterns will be generated in motion in one or several directions, so that the contrast borders cover the entire surface of the object.
    As mentioned, indirect lighting allows great flexibility in the design of patterns, both in their shape and in the size of the grid and their colors. For example, image sequences can be used in which the light pattern, instead of imitating the light stripes, becomes similar to a chess board (Figure 5) or a grid of squares (Figure 6).
    5
    10
    fifteen
    twenty
    25
    30
    35
    The objective of using this type of pattern is to minimize the number of images necessary to analyze the entire surface of the product, maximizing the amplifications of the defects in the images thanks to indirect light from all possible directions on the defects, guaranteeing the highlight of the defect regardless of the positioning of the camera, of the dominant direction of the contrast stripes of the lighting, of the direction of movement of these stripes by movement of the inspected product, or of the lighting patterns.
    Also, depending on the color of the paint, it may be necessary to locate the defects in the light or dark areas of the images registered by the cameras (2). The pattern design can also be adapted to this fact, for example the pattern shown in Figure 6 is designed to work in a dark area. An inverse pattern to this would be adequate if you want to work in a clear area of the images.
    Through this lighting system based on projection or retro-projection, homogeneous patterns of different colors and gray levels can also be interleaved, which optimize the detection of contrast defects using the same cameras (2) with which the defects of geometry.
    The procedure for detecting surface defects associated with the system of the invention is described below. First, the surfaces of the objects to be inspected are illuminated with the light-emitting means (1) and the image capture by the cameras (2) is performed (in Figure 7, as an example, a pattern obtained for an LEDS tunnel as shown in Figure 2, in a realization based on direct illumination). Orientating the cameras (2) with an angle of vision with respect to the surface to be inspected highlights the geometry defects that can be observed in consecutive images, especially around the contrast borders generated by the lighting pattern (see Figure 8).
    Once the information has been entered into the computers, they process, using algorithms and 2D filters and with the calibration of the cameras, the photogrammetric projection and the determination of the defect location during the inspection in a synchronized manner. 3D filters are also applied to the captures of the cameras (2), and only merged information is obtained of the defects on the surface to be inspected.
    The system has the possibility of adapting the capture parameters (gain, exposure time, etc.) to the specific color of the inspected product and also to keep those capture parameters updated in time and, therefore, being reactive to possible variations in the sensitivity of the cameras and the luminance of the 5 lighting systems.
    With the result of the inspection, the control and analysis subsystem can print or generate an inspection report, which will show the location of the defects, as well as their severity. Such information may also be shown as 3D information in glasses of augmented reality, for the subsequent optimized repair of these by means of the procedures that the company deems appropriate.
    权利要求:
    Claims (15)
    [1]
    5
    10
    fifteen
    twenty
    25
    30
    35
    1. - System for detecting defects in specular or semi-specular surfaces of objects to be inspected, suitable for integration into an industrial production line, comprising:
    - a plurality of light emitting means (1) on the objects;
    - a plurality of cameras (2) for the detection of the light reflected by the objects;
    - a photogrammetric control and analysis subsystem of information associated with the light emitted by the light emitting means (1) and the light detected by the cameras (2);
    - a support structure (3) in which the light emitting means (1) and the cameras (2) of the system are arranged;
    the system being characterized in that the light emitting means (1) are arranged in the system so that the light emitted on the surface of the objects has a periodic pattern of light and shadow, and because said pattern has a relative movement with respect to the objects to inspect, for the route its surface.
    [2]
    2. - System according to the previous claim, wherein the cameras (1) are capture cameras with an image acquisition frequency between 40-100 Hz.
    [3]
    3. - System according to any of the preceding claims, wherein the cameras (1) are CCD or CMOS industrial cameras.
    [4]
    4. - System according to any of the preceding claims, wherein the light emitting means (1) comprise means of direct illumination on the objects.
    [5]
    5. - System according to the preceding claim, wherein the light emitting means (2) comprise neon, fluorescent, or LED light emitters.
    [6]
    6. - System according to any of the preceding claims, wherein the light emission means (2) comprise means of indirect illumination on the objects, by means of projection or rear projection on one or more screens (4).
    [7]
    7. - System according to the previous claim, where the light emitting means (2) emit periodic patterns of light and shadow on the screens (4) which, in turn, are reflected on the objects to be inspected.
    5
    10
    fifteen
    twenty
    25
    30
    35
    [8]
    8. - System according to any of the preceding claims, wherein the relative movement existing between the periodic pattern of light / shadow and the objects to be inspected is produced by the movement of said objects, the static emitted pattern remaining on them.
    [9]
    9. - System according to any one of claims 1-7, wherein the relative movement existing between the periodic pattern of light / shadow and the objects to be inspected is produced by the movement in one or several directions of the pattern emitted on the objects to be inspected, while said objects remain static.
    [10]
    10. - System according to any of the preceding claims, wherein the information control and analysis subsystem comprises process and treatment means by photogrammetric projection of the images captured by the cameras (2), 2D and / or 3D filtering means for the identification of defects in said images, means of analyzing the color of the objects to be inspected, and / or means of representation of the objects and defects identified by augmented reality.
    [11]
    11. - System according to any of the preceding claims, which also comprises one or more of the following elements: high resolution lenses for cameras (2), one or more computers for processing and control of the system, one or more elements of measurement to control the position of the objects and synchronize the captures of the cameras (2).
    [12]
    12. - Method of identifying defects in specular or semi-specular surfaces of objects to be inspected, characterized in that it comprises the use of a system according to any of the preceding claims, and the realization of the following steps:
    - the surface of the objects to be inspected is illuminated with the light emission means (1), so that the light emitted on said surface has a periodic pattern of light and shadow, with a relative movement on the objects to be inspected;
    - the light reflected by the objects is registered by means of the cameras (2);
    - the reflected light and the light emitted through the control and analysis subsystem are analyzed, through photogrammetric projection, for the detection of surface defects.
    [13]
    13. - Method according to the previous claim, where the detection system is integrated in an industrial production line of the objects to be inspected.
    [14]
    14. - Method according to any of claims 12-13, wherein the periodic pattern of light and shadow emitted has an adaptive shape to the surface of the objects
    to be inspected, so that said pattern varies depending on the curvature of the inspected surface.
    [15]
    15. - Method according to any of claims 12-14, wherein the periodic pattern of light and shadow emitted is a checkered pattern, a grid of squares or
    triangles, or a pattern of points, straight lines, circumferences or crosses.
    FIGURES
    image 1
    FIG. one
    image2
    类似技术:
    公开号 | 公开日 | 专利标题
    ES2630736B1|2018-07-04|SYSTEM AND METHOD OF DETECTION OF DEFECTS IN SPECULAR OR SEMI-SPECULAR SURFACES THROUGH PHOTOGRAMETRIC PROJECTION
    CN107525479B|2020-03-20|Method for identifying points or regions on the surface of an object, optical sensor and storage medium
    KR101773791B1|2017-09-01|Method and device for inspecting surfaces of an examined object
    EP1649423B1|2008-08-13|Method and sytem for the three-dimensional surface reconstruction of an object
    CN101821578B|2014-03-12|Non-contact measurement apparatus and method
    US7532333B2|2009-05-12|Method and apparatus for determining the shape and the local surface normals of specular surfaces
    JP2017500540A|2017-01-05|Method and apparatus for analyzing tire condition
    US20140376003A1|2014-12-25|Arrangement for optical measurements and related method
    EP3783304A1|2021-02-24|Calibration of a triangulation sensor
    JP2010532870A|2010-10-14|Optical inspection method and inspection apparatus for object surface
    US20140002610A1|2014-01-02|Real-time 3d shape measurement system
    US20050109091A1|2005-05-26|Procedure and device for testing tires
    CN105190232A|2015-12-23|Three-dimensional coordinate scanner and method of operation
    CN105102925A|2015-11-25|Three-dimensional coordinate scanner and method of operation
    JP5051493B2|2012-10-17|3D measurement marker and 3D measurement method using the same
    TWI490445B|2015-07-01|Methods, apparatus, and machine-readable non-transitory storage media for estimating a three dimensional surface shape of an object
    JP2021527220A|2021-10-11|Methods and equipment for identifying points on complex surfaces in space
    CN105783784A|2016-07-20|Inspection device and control method of the same
    CN101086443A|2007-12-12|Device for determining precious stone shape
    JP4318579B2|2009-08-26|Surface defect inspection equipment
    US9530201B2|2016-12-27|Method for the non-destructive testing of a blade preform
    KR102357134B1|2022-01-28|Apparatus and method for inspecting tires
    JP2016166853A|2016-09-15|Location estimation device and location estimation method
    CN105960569B|2019-03-12|The method of three-dimension object is checked using two dimensional image processing
    US10151581B2|2018-12-11|Method and device for determining the position and orientation of a specular surface forming a diopter
    同族专利:
    公开号 | 公开日
    EP3388781A4|2019-07-24|
    ES2630736B1|2018-07-04|
    WO2017098071A1|2017-06-15|
    EP3388781A1|2018-10-17|
    MA43424A|2018-10-17|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    WO2021099648A1|2019-11-18|2021-05-27|Asociacion Centro Tecnologico Ceit|Dimensional control system based on photogrammetric vision|DE3712513C2|1987-04-13|1993-01-21|Daimler-Benz Aktiengesellschaft, 7000 Stuttgart, De|
    US5477268A|1991-08-08|1995-12-19|Mazda Motor Corporation|Method of and apparatus for finishing a surface of workpiece|
    US5471307A|1992-09-21|1995-11-28|Phase Shift Technology, Inc.|Sheet flatness measurement system and method|
    JP2976869B2|1995-12-28|1999-11-10|日産自動車株式会社|Surface defect inspection equipment|
    DE19944354C5|1999-09-16|2011-07-07|Häusler, Gerd, Prof. Dr., 91056|Method and device for measuring specular or transparent specimens|
    US6266138B1|1999-10-12|2001-07-24|Perceptron, Inc.|System and method for detecting defects in a surface of a workpiece|
    JP3871963B2|2002-05-16|2007-01-24|住友化学株式会社|Surface inspection apparatus and surface inspection method|
    US8705043B2|2009-12-14|2014-04-22|Academia Sinica|Height measurement by correlating intensity with position of scanning object along optical axis of a structured illumination microscope|
    US20130057678A1|2010-05-17|2013-03-07|Ford Espana S.L.|Inspection system and method of defect detection on specular surfaces|
    CN102445167A|2010-09-30|2012-05-09|旭硝子株式会社|Method and device for evaluating surface shape|JPWO2020021808A1|2018-07-26|2021-08-02|コニカミノルタ株式会社|Surface inspection equipment and surface inspection method|
    WO2020188319A1|2019-03-19|2020-09-24|Dehghanian Hosseinali|Photogrammetric augmented reality based inspection system|
    GB2582397A|2019-10-04|2020-09-23|Degould Ltd|Vehicle imaging station|
    WO2021151412A1|2020-01-27|2021-08-05|Jan Nabatian|Apparatus and method for automatically detecting damage to vehicles|
    CN111650208B|2020-06-01|2021-08-27|东华大学|Tour type woven fabric defect on-line detector|
    法律状态:
    2018-07-04| FG2A| Definitive protection|Ref document number: 2630736 Country of ref document: ES Kind code of ref document: B1 Effective date: 20180704 |
    优先权:
    申请号 | 申请日 | 专利标题
    ES201531776A|ES2630736B1|2015-12-07|2015-12-07|SYSTEM AND METHOD OF DETECTION OF DEFECTS IN SPECULAR OR SEMI-SPECULAR SURFACES THROUGH PHOTOGRAMETRIC PROJECTION|ES201531776A| ES2630736B1|2015-12-07|2015-12-07|SYSTEM AND METHOD OF DETECTION OF DEFECTS IN SPECULAR OR SEMI-SPECULAR SURFACES THROUGH PHOTOGRAMETRIC PROJECTION|
    MA043424A| MA43424A|2015-12-07|2016-12-07|SYSTEM AND METHOD FOR DETECTION OF DEFECTS IN SPECULAR OR SEMI-SPECULAR SURFACES BY PHOTOGRAMMETRIC PROJECTION|
    EP16872466.4A| EP3388781A4|2015-12-07|2016-12-07|System and method for detecting defects in specular or semi-specular surfaces by means of photogrammetric projection|
    PCT/ES2016/070865| WO2017098071A1|2015-12-07|2016-12-07|System and method for detecting defects in specular or semi-specular surfaces by means of photogrammetric projection|
    [返回顶部]