Method for inspecting coating surfaces for coating defects

Abstract

To provide a method of inspecting a coated surface for surface defects.SOLUTION: A method disclosed herein comprises: a step (102) of using a device (200, 300, 400, 700, 800, 900) for covering a coated surface to be inspected, the device being configured to create an enclosed space to isolate the surface coating to be inspected from ambient illumination in order to provide predefined photographic acquisition conditions within the enclosed space; a step (104) of acquiring a photo of the coated surface within the enclosed space; and a step (106) of inspecting the photo for the presence of surface defects.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to a method for identifying coating defects and characterizing a coating surface, particularly to a method for characterizing a coating surface based on a coating composition of paints, varnishes, printing inks, polishing resins, pigment concentrates, or other coating compositions.

Background Art

[0002] Printing and varnish coatings have various defects, which have an adverse effect on the appearance or technical properties of the coated object. Coating defects are, for example, bubbles, craters, cloudiness, leveling problems, wetting problems, pigment floaters (floating), sagging, agglomeration, or bubble formation, and some of these defects occur simultaneously and interact with each other. To evaluate and avoid these problems, the formulation is coated on a test substrate during the development of the formulation and during the inspection of defects. Different substrates are used depending on the intended application field, for example, wood, plastic, paper / cardboard, glass, or metal. Also, different pretreatments are possible for the substrate, and the pretreatment can further complicate the problem.

[0003] Currently, defects are visually evaluated by a person, such as an employee. This purely visual evaluation is usually very rough, subjective, and has little reproducibility. As a result, the identification of defects and the evaluation of the quality of the coating surface require a lot of experience on the part of the employees, and they vary greatly depending on them, so it is difficult to compare the results. For example, the test substrate is irradiated using a table lamp or sunlight shining in through a window. The coating defects determined manually in this way can hardly be compared due to the variation of different environmental parameters.

Summary of the Invention

Problems to be Solved by the Invention

[0004] The object of the present invention is to provide an improved method and apparatus for inspecting a coated surface for surface defects, as described in the independent claims. Embodiments of the present invention are described in the dependent claims. Embodiments of the present invention can be freely combined with each other if they are not mutually exclusive. [Means for solving the problem]

[0005] In one aspect, the present invention relates to a method for inspecting a coated surface for surface defects. The method is: A step of using an apparatus to cover a coating surface to be inspected, wherein the apparatus forms a closed space, isolates the coating surface to be inspected from ambient light, and provides predetermined photographic acquisition conditions within the closed space. The steps include: obtaining a photograph of the coating surface located within the enclosed space; The steps include: inspecting the photograph to determine whether or not there are surface defects; It holds.

[0006] This can be significant because the device ensures that the surface is illuminated in a predetermined and reproducible manner, and the photographs of coating defects provide reproducible results that are compatible with results obtained from other photographs regarding the type and number of coating defects observed. Therefore, the device suppresses variations in illumination intensity or illumination angle that are unavoidable when using, for example, daylight or a table lamp. For example, the incidence of daylight through a window depends on the time of day, season, and weather. When using a table lamp, there is a risk that the lamp may move intentionally or accidentally, the work table may be powered on, or the lamp holder may rotate. Also, the object being photographed placed near the coated surface may cause light reflection, resulting in the surface being illuminated by an additional light source from time to time. In contrast, the use of the device makes it possible to protect and isolate the coated surface from ambient light and provide predetermined photographic acquisition conditions within a closed space.

[0007] It is known that the conditions under which photographs are taken have a significant impact on the detection results of coating defects. This is especially true for very small coating defects, such as microbubbles. Due to the small size of the defects, unfavorable photographic conditions may cause the defects to become unrecognizable in the image. In embodiments of the present invention, photographs are taken under optimal or at least sufficiently favorable conditions, allowing for consistent conditions and improved compatibility of results.

[0008] Optimal and consistent photographic conditions can be achieved by the apparatus described in relation to embodiments of the present invention. The apparatus may have one or more photographic components (light source, camera, sample carrier) for acquiring photographs, thereby the apparatus components are coordinated with each other. Since different coating defect types are observed by applicants requiring different photographic conditions, the components are adjusted (automatically, if necessary) to suit each type and calibrated, if necessary.

[0009] In embodiments of the present invention, the coating surface can be shielded from the influence of ambient light, ensuring certain defined photographic conditions, reproducible image analysis results, and reproducible defect identification results. This is particularly beneficial when the photographs are automatically inspected by image analysis software and / or used as training images to run a machine learning method that creates a predictive model for detecting and / or characterizing coating defects.

[0010] In one embodiment, the apparatus has a housing that is opaque to light, and the use of the apparatus includes the step of covering the coating surface, for example, the use of the apparatus may include the step of covering the coating surface with the housing of the apparatus. In one embodiment, the housing has an opening side, and the use of the apparatus includes the step of bringing the opening side of the housing into contact with the coating surface, and / or The aforementioned device is Having one or more light sources, and / or The device has a camera housing opening that allows images of the sample to be taken with at least one in-device camera and / or at least one external camera, and the step of taking the photograph is performed with the at least one camera.

[0011] In one embodiment, the device further The use of a device that has a sample carrier for transporting a coating sample including a coating surface, and covers the coating surface, includes the step of placing the sample on the sample carrier, and / or The coating surface has an opening, the coating surface opening being an opening in the wall of the housing referred to as a contact wall, the contact wall being a wall of the housing intended to be in contact with the coating surface, and / or the use of the device covering the coating surface has a step of bringing the contact wall into contact with the coating surface. The device has a sample position marking, which indicates the position within the apparatus where the sample having a coated surface is placed.

[0012] A "camera housing opening" is an opening that is fitted (e.g., dimensionally and shaped) so that at least one external camera can take an image of a sample through the camera housing opening. For example, the external camera is the camera of a portable communication device positioned on the outer surface of the housing wall having the camera housing opening, and the communication device may fully integrate the camera housing opening. This allows the device housing and the communication device positioned on the camera housing opening to together isolate the surface coating inside the device from ambient light. In one embodiment, the camera housing opening is resizable. For example, the camera housing opening may have one or more sliding doors or panels, and the size of the camera housing opening may be changed. This has the advantage that the camera housing opening can be fitted to smartphones or document cameras of different sizes.

[0013] In one embodiment, the photograph is acquired in the form of a digital image showing the coated surface of the sample.

[0014] The use of a device having at least one light source may have the advantage that the intensity of the light source, the position of each light source, and / or the irradiation angle can have defined, reproducible characteristics. In a preferred embodiment, at least one light source is a light source having a vertical and / or horizontal arrangement within the device, and its intensity, color spectrum, and / or irradiation angle are controllable and adjustable, so that a photograph of the coating surface can be taken at a predetermined irradiation intensity, irradiation spectrum, light source position, and / or irradiation angle.

[0015] The use of a sample carrier may have the advantage that a sample having a coated surface can have a predetermined, reproducible position relative to the housing of the apparatus. In a preferred embodiment, the sample carrier is one whose height and / or position within the apparatus can be adjusted manually or automatically. This allows for the capture of a photograph of the coated surface at a predetermined distance from at least one light source and at least one camera.

[0016] Embodiments in which the housing includes a coating surface opening may have the advantage of allowing the device to be used with extreme flexibility. For example, the device may have a cubic or cylindrical shape, which has the "coating surface opening" on the (preferably flat) housing surface. Here, the device can be in contact with any coating surface. For example, the coating surface may be a horizontally oriented surface, such as the top surface of a table, and the device may simply be placed on the table surface with its "contact side" facing downwards (the contact side may also be the "bottom side"). In other examples, the coating surface may be a vertically oriented surface, such as a coated window or a coated vehicle door. The device may simply be pressed against the vertically oriented surface, with its contact side having the "coating surface opening," and may be held in this manner during image acquisition. Preferably, the coating surface opening is smaller than the area of ​​the coating surface, and the coating surface is substantially flat. This allows the device to fit snugly against the coating surface, so that little or no ambient light can enter the interior of the device.

[0017] The use of a device having at least one in-device camera may have the advantage that the position and / or image acquisition angle of each camera may have defined, reproducible characteristics. In a preferred embodiment, the at least one in-device camera is a camera whose vertical and / or horizontal position and / or orientation within the device, as well as its image acquisition angle, are controllable and adjustable. This allows a photograph of the coating surface to be taken at a predetermined image acquisition angle and / or a predetermined distance.

[0018] The use of a camera housing opening may have the advantage of allowing the use of widely available cameras, such as the cameras of communication devices like smartphones, when acquiring photographs of the coated surface. Since the communication device needs to be positioned in the same location at all times, the camera of the communication device is ensured to be positioned above the camera housing opening, and the distance between the camera and the coated surface is ensured to be a predetermined, repeatable distance.

[0019] In some embodiments, the method includes the step of storing the acquired digital images in a storage medium. For example, the storage medium may be a data storage medium provided as a component of the device. Alternatively, the storage medium may be a storage medium of a portable communication device, or a portable data carrier such as a chip card or USB stick, which may be operably coupled to or capable of coupling with the device's controller module. In addition to or separately from this, the images are stored on a remote server computer. For example, the device may have an interface that operably couples with a device having a controller module operating on an external data processing device, such as a portable communication device. The interface may be configured to receive control commands to the controller module and / or transmit device status information and / or acquired images to the controller module.

[0020] In some embodiments, the method includes the step of transferring acquired digital images to image analysis software configured to perform image analysis. For example, the image analysis software can be provided as an integrated part of the device, or run on a portable communication device interoperable with the device, and / or run on a server-computer system interoperable with the device or communication device.

[0021] The step of running image analysis software on a communication device or server typically has the advantage that the server and communication device have greater computing power than the control unit provided as a component of the device.

[0022] In one embodiment, the apparatus has a controller module and / or is operably coupled to a controller module. The controller module may be, for example, an integrated part of the apparatus or an external member of the apparatus, whereby the apparatus has an interface for exchanging data with an external controller module of the apparatus. The controller module is configured to control the operation of the apparatus, which operates and acquires a photograph only in a defined coating defect type identification mode. For example, the controller module can be based on software, firmware, hardware, or a combination of two or more of the foregoing items.

[0023] In one embodiment, the method further comprises providing a configuration, the configuration including a plurality of assignments, each assignment assigning one of a plurality of different coating defect types to one of a plurality of apparatus settings, each apparatus setting determining one or more apparatus configuration parameters that specify the position and / or mode of operation of one or more members of the apparatus, receiving, by the controller module, a selection of at least one of the coating defect types, identifying, by the controller module, an apparatus setting stored in the configuration in association with the selected coating defect type, modifying, by the controller module, at least one member of the apparatus, the at least one member being selected from the group including a camera, a light source, and / or a sample carrier according to the identified apparatus setting, whereby the mode of operation of the apparatus is set to a defect type identification image acquisition mode, and having.

[0024] A photograph is acquired while the apparatus is in the defect type identification image acquisition mode.

[0025] For example, the configuration may consist of one or more configuration files, or it may include one or more data records stored in a database.

[0026] This has the advantage that the device can operate in one of several predetermined coating defect type-specific operating modes. Each operating mode may have a predetermined selection of a light source and / or camera used to illuminate the sample and acquire a photograph. This ensures that a specific operating mode is set for a particular type of coating defect, thereby ensuring that there are optimal or at least significantly preferred image acquisition conditions for each defect type, and that a human inspector or image analysis program can easily detect and / or quantify the defect of that defect type. In some embodiments, the device has an interface, and by another component, such as a controller module, the device can switch from one defect type-specific operating mode to another. The operating state of the device may be changed to a different operating mode in response to a user selection of one or more coating defect types to be detected. The step of setting the device to a particular defect type-specific operating mode may include setting the state of one or more components of the device (e.g., one or more properties such as position, orientation, and intensity) according to the device settings specified in the configuration for the selected defect type. The applicant has observed that environmental parameters suitable or optimal for detecting and / or quantifying a given defect type, such as light intensity, angle of incidence, and distance from the sample to the camera, can vary significantly for different defect types. By adapting the operating mode of the apparatus to the selected defect type, it is possible to ensure that conditions suitable for each defect type are present during photography. This embodiment may be particularly advantageous for apparatus components that can automatically set and change position, orientation, and / or other parameters, such as a robotic arm or vertical and / or horizontal transport rails used to move the camera and / or light source, or an automatic lift to change the position of the sample.

[0027] In some embodiments, the device has multiple light sources. When the device operates in a first operating mode, one or more first light sources are selectively activated. The first operating mode is specified via a first device setting. The first device setting is suitable for taking photographs that can detect and / or quantify a first type of coating defect. One or more second light sources are selectively activated when the device operates in a second operation. The second operating mode is specified via a second device setting. The second device setting is suitable for taking photographs that can detect and / or quantify a second type of coating defect. Thus, when a first type of coating defect is selected, only one or more first light sources are activated and the second light sources are turned off. When a second type of coating defect is selected, only one or more second light sources are activated and one or more first light sources are turned off.

[0028] This may have the advantage of not requiring the motor to be integrated with moving or rotating parts in order to change the position and / or orientation of the light source. The light source may already be mounted inside the housing of the device, and a subset of the light source may already be positioned and oriented so as to illuminate the coating surface in a manner suitable for detecting one or more defect types.

[0029] In one embodiment, the method further, - A step of providing a configuration, wherein the configuration includes a plurality of assignments, each assignment assigning one of a plurality of different coating defect types to one of a plurality of device settings, and each device setting determining one or more device configuration parameters that specify the position and / or mode of operation of one or more components of the device, - The controller module receives a selection of at least one type of coating defect, - The controller module identifies the device settings stored in the configuration in association with the selected coating defect type, - The controller module outputs a message to the user, the message representing a method for modifying the position, orientation, and / or mode of operation of at least one component of the apparatus according to the identified apparatus settings, wherein the at least one component is selected from a group including a camera, a light source, and / or a sample carrier, and the message enables the user to manually or semi-automatically set the mode of operation of the apparatus to a defect type identification image acquisition mode. It holds.

[0030] The acquisition of the digital image is performed while the device is in the defect type identification image acquisition mode.

[0031] This embodiment allows for the inexpensive manufacture of the apparatus, as individual components such as sample carriers or light sources do not require motors or other means for automatically changing their position or state, and may be particularly advantageous in terms of robustness. Any necessary modifications are simply indicated, and the user can manually make appropriate adjustments to the various components of the apparatus. Preferably, manual adjustment means such as knobs, rails, and joints have markings or labels, so that the user can adjust the position, orientation, and / or state of each component at a separate point.

[0032] For example, the sample carrier may be a manually adjustable lift with manually movable wheels, allowing for adjustment of the lift's height. In some examples, separate markings are displayed on the wheels to indicate a predetermined number (e.g., five) of different lift positions and corresponding heights. When a user selects a specific defect type, a controller module, for example, which is part of the apparatus or operates on the user's portable communication device, determines that the lift should have a predetermined position specifically assigned to the selected defect type. For example, the lift positions may be stored in configurations associated with each defect type. For example, the controller may determine that the lift should have the highest position "5" to bring the sample closer to the camera. Via the GUI, the user is notified that the sample carrier's wheel needs to be rotated until one of the marks on the wheel representing a predetermined position (e.g., the "5" mark on the wheel) reaches the target position.

[0033] In one embodiment, one or more device configuration parameters that can be adapted to change the operating mode of the device are: - Identification of each of one or more light sources included in the device to be turned on. In some examples, the light source has a light panel, and the controller may be configured to turn on the light panel when one or more predetermined first defect types, such as "translucent defect," are selected; - Identification of each of one or more light sources included in the device to be turned off. In some examples, the light sources may have a light panel. The control controller may be configured to turn off the light panel when one or more predetermined second defect types, such as "crater defects," are selected; - The light intensity of one or more of the multiple light sources included in the apparatus; - The position of each of one or more light sources of the plurality of light sources included in the apparatus, in particular the height and / or horizontal offset of each light source with respect to one or more of the walls of the housing; - The distance of each of the one or more light sources of the multiple light sources included in the apparatus to the coating sample; - The irradiation angle between each of the one or more light sources of the plurality of light sources included in the apparatus and the coating sample; - The position and / or height and / or inclination of the sample carrier within the apparatus. The “inclination” of the sample carrier may be, for example, the inclination of the plate of the sample carrier, which is designed to transport the sample placed on the plate. For example, the sample carrier may have one or more joints that allow the plate to pivot in one or more directions. Preferably, the sample carrier is designed so that pivoting is possible only with respect to a limited number of preset inclination angles (inclination angles of the sample carrier, in particular the carrier plate); - Identification of one or more in-device cameras included in the apparatus described in claim 2, wherein the in-device cameras are used for image acquisition; - Identification of one or more in-device cameras included in the apparatus according to claim 2, wherein the in-device cameras are not used for image acquisition; - The arrangement of one or more of the multiple in-device cameras included in the apparatus described in the embodiment of the present invention, in particular, the height and / or horizontal offset of each camera relative to one or more walls of the housing; - The distance of one or more of the multiple cameras included in the apparatus to the coating sample; - The image acquisition angle between each of the multiple cameras included in the apparatus and the coating sample; It is selected from the group that includes it.

[0034] The use of a light panel can be particularly significant in identifying coating defects that affect the transparency (or translucency or opacity) of the coating surface, or the uniformity of the transparency (or translucency or opacity) distribution. For example, if the coating is applied by spraying from a spray nozzle and the nozzle has some defect, or if the coating composition contains debris or other types of solid or semi-solid contaminants that could partially clog the nozzle, coating defects such as a coating that is too thick, too thin, or an uneven distribution of the amount of coating applied to the surface may occur. If too much coating is applied, most of the light emitted by the light panel will be absorbed by the coating surface, and the transparency (or translucency) shown in the acquired image and determined by the image analysis software may fall below a predetermined minimum transparency (transparency threshold). If an insufficient amount of coating is applied, the transparency (or translucency) shown in the acquired image and determined by the image analysis software may exceed a predetermined maximum transparency (translucency threshold). If the nozzle becomes clogged with contaminants contained in the coating composition, the observed distribution of transparency (transparency) on the coating surface may become non-uniform.

[0035] In one embodiment, one or more of the aforementioned components can have their respective configuration parameters adjusted to one of a limited number of predetermined parameter values. For example, the device components may have a mechanical control element such as a wheel or switch, or an electronic interface that allows a user or controller module to set the parameter value to one of a predetermined number of acceptable values. For example, two, three, four, five or more light sources or cameras may be positioned in predetermined locations. Also, the angle, light intensity, sample carrier height, etc., may preferably be set to one of a predetermined limited number of possible values. This ensures that the illumination conditions used to obtain photographs of a specific type of coating defect are compatible.

[0036] In some embodiments, the apparatus has a sample carrier configured to transport a sample having a coated surface to be inspected, the sample carrier being able to be manually or automatically adjusted in height and / or position within the apparatus. In some examples, the apparatus may further have a light panel positioned on or integrated with the sample carrier.

[0037] The use of a light panel in combination with a height-adjustable sample carrier has been observed to be particularly significant in identifying coating defects related to the application of inappropriate amounts of coating composition (e.g., excessive or insufficient amounts of coating composition, or uneven distribution of coating composition). This is because the light intensity captured by the camera is used to measure the transparency (or transmittance or opacity) of the coating, and the light intensity also depends on the distance between the camera and the coating surface. Therefore, using a height-adjustable sample carrier in combination with a light panel can provide particularly accurate identification and characterization of the aforementioned type of coating defect by image analysis software. In some examples, the light panel is positioned on top of the sample carrier or is an integrated part of the sample carrier (e.g., the top plate).

[0038] In one embodiment, the device has a display configured to show a GUI generated by a controller module. For example, the display may be an LCD or OLED display integrated into or mounted on the housing of the device.

[0039] In another embodiment, the device has and / or is operably coupled to a portable communication device. For example, the device may include a near-field communication interface such as a Bluetooth® interface, or a network interface for exchanging data with the communication device over a network, such as the Internet. The portable communication device has a controller module. For example, the controller module may be implemented in the form of an application program, particularly a so-called "app". The portable communication device has a display, and the controller module may be configured to generate a GUI.

[0040] A GUI, displayed via the device's display and / or the portable communication device's display, allows the user to select one or more defect types of interest, control the operating mode of the device and / or individual components of the device, and / or view the results of the detection and / or quantification of one or more coating defects.

[0041] In one embodiment, the apparatus has an interface for exporting acquired photographs of coating samples. For example, the export interface may be the aforementioned near-field communication interface, network interface, or another interface, such as a USB interface or chip card interface.

[0042] In one embodiment, photographic inspection is performed by image analysis software. The image analysis software may be installed on a data processing unit, which is an integrated part of the device. In addition to or separately from this, the image analysis software may run on a portable communication device, a personal computer, or a server operably coupled to a portable communication device or personal computer. In some embodiments, the controller module and the image analysis software are part of the same monolithic software (standalone software application) or part of a distributed software framework.

[0043] In one embodiment, the method further comprises the step of forming a graphical user interface (GUI) using a controller module. The GUI is configured to display a plurality of selectable GUI elements, each representing one of the plurality of different coating defect types.

[0044] The controller module receives at least one selection of the coating defect type in the form of a user's selection of each of the GUI elements.

[0045] This is significant as an extremely flexible method, providing a system in which the user can select the defect type of interest via a GUI. The selection can, for example, trigger the display of automatic adjustment of the device's operating mode and / or necessary manual control steps to adjust the device's operating mode, and a photograph can be taken under conditions suitable for detecting the selected defect type.

[0046] In one embodiment, this method is - A step of acquiring a digital preview image using at least one camera, wherein the preview image shows the coating surface, - A step of performing an initial image analysis of the preview image using the controller module, wherein at least one type of coating defect shown in the preview image is identified. It has the following characteristics: At least one selected coating defect type is at least one coating defect type identified in the initial image analysis.

[0047] For example, preview images can be captured in the device's preview mode. Its primary and typical purpose is to determine, in a first rough automatic or manual image analysis step, which defect types are present on the coating surface, or at least present with a minimum probability. Preferably, the preview mode allows for the detection of as many different defect types as possible. In one embodiment, a list of (possible) defect types determined in the preview step is displayed to the user for selecting one or more defect types, or is used to automatically set the mode of operation of the device to a set of modes suitable for taking photographs. This allows for the detection and / or quantification of each defect type.

[0048] In one embodiment, upon receiving a selection of multiple defect types, the controller module causes the device to enter a sequence of multiple operating modes, in which a photograph of the coating surface is taken for subsequent inspection by a person or one or more image analysis algorithms.

[0049] In one embodiment, the photographic inspection is performed by image analysis software and includes the step of obtaining qualitative and / or quantitative characterization of coating defects on the coating surface. The step of obtaining qualitative and / or quantitative characterization of coating defects on the coating surface may be performed separately for one or more defect types of interest, for example, in parallel or sequentially. The method further includes the step of outputting the results of the image analysis by image analysis software.

[0050] In one embodiment, the apparatus has a plurality of light sources. Preferably, one of the plurality of light sources is a light panel. The light panel can be permanently installed in the apparatus or simply inserted into the apparatus so that it can be removed at any time. For example, the light panel can be located at the bottom of the apparatus or on the carrier plate of the sample carrier. Preferably, the apparatus can have an opening for the power lines of the light panel, which is large enough for the power cables to fit therein, so that the cables are not compressed on the one hand and excessive stray light does not enter on the other hand.

[0051] A light panel represents the sample support plate of a sample carrier, or a light panel placed on top of a sample carrier. The light panel functions as a transmitted light source, emitting light that passes through the coating surface before it is acquired by a camera that forms a photograph.

[0052] This embodiment may be particularly significant in detecting coating defects within a coating where differences in coating thickness are considered particularly undesirable. This is the case, for example, with coatings applied to transparent surfaces, or with protective coatings where a minimum thickness needs to be ensured throughout the coating to provide a protective effect. Preferably, the sample covered with the surface coating is a transparent sample, such as a glass slide.

[0053] For example, the light panel may be a light source in the form of a flat sheet of material. Preferably, the light panel is larger than the size of the sample that can be placed on top of the light panel. This ensures that all areas of the coating sample surface are uniformly illuminated by the light from the light panel. For example, the light panel may have a square or rectangular shape, with each side of the light panel ranging from, for example, 5 cm to 40 cm, or from 5 cm to 20 cm. The light panel can be an LED light panel.

[0054] In one embodiment, at least one light source has multiple light sources.

[0055] In some embodiments, each of at least a subset of light sources may be individually controlled with respect to the on state, off state, and / or light intensity. This means that a controller module may be configured to individually turn each light source in the subset on or off, and / or to individually set the brightness of the light sources according to, for example, a select defect type of the device or a specific operating mode.

[0056] This has the advantage of allowing different coating defects to be detected by setting different illumination conditions, without needing to change the position or orientation of the coating surface and / or the camera. This can be achieved by simply switching different light sources on and off or increasing or decreasing the brightness of some light sources. On the one hand, since no movement of the sample or mechanical device components is required, this allows for switching between device operating modes in a shorter time. Therefore, the time required to acquire images of different defect types under different conditions can be significantly accelerated. Also, in some cases, the device does not need to have moving parts, resulting in a mechanically robust device that is particularly suitable for mobile applications.

[0057] In some embodiments, at least one of the light sources is a ring-shaped irradiation source. This may have the advantage that the coating surface can be irradiated more uniformly.

[0058] In some embodiments, at least one of the light sources is configured to adjust the size of the optical cone to a limited number of predetermined shapes. For example, the light source may have a wheel or switch or an electronic interface that can focus the beam emitted by the light source to a greater or lesser degree at a point. For example, two, three, four, five or more predetermined selectable optical cone shapes may be supported by the light source.

[0059] In one embodiment, the device is a single component and / or a portable device. Preferably, the portable device has a weight of less than 30 kg, preferably less than 20 kg, or less than 10 kg, and preferably has dimensions that allow a person to move the box within a short range of at least 3 to 4 meters, and has a holding device.

[0060] In one embodiment, a photographic inspection for the presence or absence of surface defects includes the step of identifying and / or quantifying one or more coating defect types. The coating defect types are selected from the group including crater defects, wear defects, poor adhesion defects, alligator defects, bleeding defects, blister defects, bloom defects, bridging defects, bubbling defects, cathodic decomposition defects, check defects, repellency defects, spiderweb defects, crack defects, fissure defects, crossfoot defects, delamination defects, fade defects, flaking defects, greening defects, thermal defects, impact defects, intercoating defects, mud crack defects, orange peeling defects, peeling defects, pinhole defects, ripple coating defects, run defects, rust spray defects, rust spot defects, rust stain defects, sagging defects, settling defects, skinning defects, solvent detachment defects, solvent splash defects, stress crack defects, undercut defects, and wrinkle defects.

[0061] In one embodiment, the apparatus is used to obtain multiple photographs of multiple coated surfaces. The method further includes, - A step of storing the acquired photograph as a digital image associated with an image label in a storage medium, wherein the image label represents the location and type of one or more surface defects of one or more different coating defect types, - A step of inputting the acquired digital images and their associated labels into a machine learning program, wherein the machine learning program is configured to perform a machine learning method that learns the relationships between the image features of the acquired digital images and the types and / or quantities of the image defects indicated in the labels, and the learned relationships are stored as a trained predictive model, - A step of integrating the predictive model into image analysis software, wherein the image analysis software is configured to qualitatively and / or quantitatively characterize defects on the coating surface of a sample shown in a digital image provided as input, It holds.

[0062] This can be beneficial in that it can improve the quality of the training dataset and, furthermore, the accuracy of the predictive model. All training images were taken under defined photographic conditions that are favorable for identifying and characterizing specific annotated defect types. The number of training images required to obtain the minimum necessary accuracy in identifying and / or quantifying coating defects (and consequently the amount of storage space, time, and CPU capacity required for training) can be reduced.

[0063] In one embodiment, image inspection is performed by image analysis software including a predictive model, so that the coating defect type-specific mode of operation of the device used when acquiring a photograph of the coating surface is equal to the coating defect type-specific mode of operation of the device used to acquire training images for the predictive model in the image analysis software. This ensures that the illumination conditions when taking the photograph of the coating surface under inspection are matched to, or at least similar to, the illumination conditions when taking the photograph used as training data for the model in the image analysis system.

[0064] The apparatus and / or system described in this application can be used to acquire images to be analyzed in order to characterize the quality of a coating composition and / or the quality of a coating process. This is described, for example, in three European patent applications with application numbers 2019660.3, 21196972.0 and 21196973.8, which are incorporated in their entirety by reference. As described in the three European patent applications mentioned above, the images can be used as training images to form annotated image analysis software configured to identify and / or characterize coating defects.

[0065] In another aspect, the present invention relates to an apparatus configured to form a closed space that isolates a surface coating from ambient light and to provide predetermined photographic acquisition conditions within the closed space. The apparatus is A housing that is opaque to light, and / or One or more light sources, and / or At least one in-device camera and / or camera housing opening within the housing, wherein the camera housing opening allows at least one external camera to acquire an image of the sample through the camera housing opening. Any sample carrier for transporting coated samples, and / or coating surface openings, and / or sample position markings, It has, The aforementioned coating surface opening is an opening within the wall of the housing, referred to as a contact wall, and the contact wall is intended for flat contact of the coating surface. The marking indicates the position within the apparatus where the sample having the coated surface is placed.

[0066] In another aspect, the present invention relates to a system having the apparatus described in an embodiment of the present invention, wherein the system is A controller module configured to control the operation of the device, and / or A configuration comprising multiple assignments, each assignment assigning one of several different coating defect types to one of several device settings, each device setting determining one or more device configuration parameters that specify the position and / or mode of operation of one or more components of the device, and / or A computer system operably coupled to the controller module, the computer system including image analysis software configured to perform image analysis to identify and / or quantify one or more coating defect types on the coating surface shown in the photograph, the computer system being particularly a server computer or a portable communication device, It holds.

[0067] The embodiments and examples described herein are understood to be illustrative examples of the present invention. Other embodiments of the present invention are envisioned. Although the present invention is described as an example of a particular combination and distribution of software programs and computer systems, it is understood that any feature described with respect to any one embodiment may be used alone or in combination with other described features, or in combination with one or more features of any other embodiment, or in any combination of any other embodiment, provided that the features are not mutually exclusive.

[0068] A "photograph" (also known as an "image" or "picture") is an image formed by light entering a photosensitive surface, typically photographic film or an electronic image sensor, such as a CCD or CMOS chip. Photographs are formed using a camera, which uses a lens to focus the wavelengths of visible light in a scene onto a replica of what is visible to the human eye. Photographs can be digital images or analog hard copies, such as printouts.

[0069] As used in this application, “apparatus” refers to physical hardware or equipment designed for a specific purpose. For example, the purpose is to provide a closed space that isolates a surface coating under inspection from ambient light and to provide predetermined photographic acquisition conditions within the closed space. The apparatus may have one or more components, such as a camera or light source, which can contribute to providing predetermined photographic acquisition conditions. The apparatus has a housing, and various components (camera, light source, sample carrier, handle for transporting the apparatus) can be mounted inside or outside the housing. The apparatus may be a single unit of hardware or equipment. The apparatus may be a single piece or a combination of multiple pieces adapted and molded to form a closed space, for example, to isolate a surface coating under inspection from ambient light and to provide predetermined photographic acquisition conditions within the closed space. In some embodiments, the apparatus may essentially provide complete isolation from ambient light. In other embodiments, the apparatus tolerates a certain amount of stray light. For example, when an external camera is positioned above a camera housing opening, it has been observed that small amounts of stray light that may enter through a small light slit in the camera housing opening are generally not harmful. This is especially true when the intensity of the internal light source is sufficiently high and the stray light is negligible relative to the intensity of the light source.

[0070] As used in this application, “housing” refers to a single or multiple pieces of material, or a combination of materials, that form the outer wall of the device. The housing may have a rectangular parallelepiped, a cylinder, a cube, or any other 3D shape. Preferably, the housing has at least one basically flat wall which can contact a flat coating surface or allow a coating sample to be stably placed inside the flat wall. The housing may form and surround a cavity and may have one or more openings which can be closed to shield the cavity from ambient light. The cavity may also be formed by a combination of the housing and one or more other objects which are, for example, portable communication devices, such as a smartphone, which are placed in an opening in the housing and which essentially prevent ambient light from entering the cavity through the opening.

[0071] In some embodiments, the housing may have, or be entirely composed of, a rigid material such as plastic, wood, plywood, metal, fiberglass, or composite material. Alternatively, parts of the housing may be made of flexible material. For example, parts of one or more walls of the housing may include, or be composed of, flexible material such as woven fabric or flexible polymer foil. This can help reduce the weight of the device.

[0072] In some embodiments, the apparatus has a frame and a cover, which are used to isolate the coated surface from ambient light. The “cover” as used in this application is a flexible, sheet-like opaque material, such as a woven fabric, fiber mesh, or flexible polymer foil, which is stretched or draped across the frame to form a closed space.

[0073] As used in this application, "camera" refers to an optical instrument that captures visual images. At a basic level, a camera consists of a sealed box (camera body) with a small hole (aperture) through which light can capture an image on a photosensitive surface (usually photographic film or a digital sensor).

[0074] As used in this application, “controller module” is a piece of software, firmware, or hardware, or a combination thereof, configured to control another article, such as a device and / or the operation of the device.

[0075] As used in this application, the “coated surface” is the surface of a substrate that has been coated with the coating composition at least partially once or multiple times. For example, the coating composition can be applied by coating, spraying, or painting the coating composition onto the substrate, by immersing at least one surface of the substrate in the coating composition, or by other coating approaches. The substrate may be any material, such as wood, glass, metal, or plastic. The substrate may be, for example, a sample or a workpiece.

[0076] As used in this application, “sample” refers to a material sample. Samples are typically small in size and can be completely placed within a confined space formed by the apparatus. In particular, a sample may be a small piece of coated material for the purpose of evaluating the quality of the coating composition and / or coating process.

[0077] As used in this application, “coating defect” or “coating surface defect” is any optically detectable deviation of the coating surface from the typical or desired appearance of the coating surface. In particular, coating defects may result from bubbles, cracks that occur during the drying process, an uneven distribution of the amount of coating material on the surface, or an uneven distribution of individual components within the coating.

[0078] As used in this application, “program” means a piece of software, such as an application program, or a module or function of an application program, or a script, or any other kind of software code executable by one or more processors, such as a CPU or GPU. A program may be, for example, an application program. An application program, in particular a program designed to run on a portable device, may be implemented in the form of an “app”.

[0079] Image analysis software used in this application for inspecting a coated surface for surface defects is a software program or software module configured to analyze a digital image to automatically identify the identity and / or location and / or quantity of one or more coating defects of one or more coating defect types. The image analysis software may also be configured to calculate a coating quality report as a function of the identified coating defects on the surface shown in the image.

[0080] As used in this application, “model” or “predictive model” is a data structure or an executable software program or program module configured to generate predictions as a function of input data. For example, a model may be a model obtained in a machine learning process by training the model on manually and / or automatically annotated training data. Alternatively, a predictive model as used in this application may include a set of two or more predictive models, for example, a set of multiple defect type identification predictive models.

[0081] Hereinafter, embodiments of the present invention, which are merely illustrative, will be described in detail with reference to the drawings. [Brief explanation of the drawing]

[0082] [Figure 1] This is a flowchart showing a method for inspecting a coated surface. [Figure 2] This figure shows an apparatus for coating a surface according to one embodiment. [Figure 3] This figure shows an apparatus for coating a surface according to another embodiment. [Figure 4] This figure shows an apparatus for coating a surface according to yet another embodiment. [Figure 5] This figure shows two photographs of the same surface taken under different environmental parameters. [Figure 6] These are two photographs of the same surface, one of which has a label indicating a surface defect. [Figure 7] This figure shows an apparatus for coating a surface according to one embodiment. [Figure 8] This is a diagram showing a system that includes devices and communication equipment. [Figure 9] This diagram shows a system comprising a server, multiple devices, and multiple communication devices. [Figure 10] This diagram shows a GUI that allows the user to select one or more defect types. [Modes for carrying out the invention]

[0083] Figure 1 is a flowchart of a method for inspecting a coated surface 204. This method can be carried out using the apparatus described in this application for embodiments of the present invention, for example, the apparatus shown in Figures 2 to 4 and Figures 7 to 9. The method has a first step 102 using apparatus 200, 300, 400, 700, 800, 900 for coating the coated surface to be inspected. The apparatus forms a sealed space, isolating the coated surface to be inspected from ambient light, and provides predetermined, preferably defect type, photographic acquisition conditions within the sealed space. Next, in step 104, a photograph of the coated surface is taken using the apparatus. The photograph may be a digital image or a hard copy photograph, for example, a paper-based printout. Next, in step 106, the photograph is inspected manually or automatically by image analysis software for detecting the presence of surface defects.

[0084] Figure 2 shows a coating apparatus 200 according to one embodiment of the present invention. The apparatus shown in Figure 2 is a cubic apparatus having a housing 218 enclosing six walls. The housing of the apparatus may be made of any material substantially opaque to light, such as cardboard, plastic, textile, metal, composite material, or a combination thereof. The apparatus has at least one opening into which a coating sample 204 is inserted. Although not shown in Figure 2, the opening may be implemented as, for example, a door or flap. In one embodiment, the inner walls of the apparatus are dark in color, such as black, to minimize reflection.

[0085] Accordingly, a sealed space that isolates the surface coating from ambient light may be provided by the housing of the device. Predetermined photographic acquisition conditions may be provided by the housing, one or more light sources, and one or more cameras having predetermined positions and / or orientations. The device may have various components that can be mounted, for example, inside the ceiling 214 or on the side walls 216 of the housing 218. For example, the components may include one or more light sources 208, such as LEDs. The components may also have at least one in-device camera 202. A lifting platform, also referred to as a “lift” 210, is positioned on the floor 212 of the device 200. The lifting platform has a support plate 206 that can move up and down vertically. For example, a manually operated or automated actuator, such as a wheel or motor, may be used for this purpose. In some embodiments, the components and / or the device 200 have means for manually or automatically adjusting the position and / or orientation of the light sources and / or cameras, and the means preferably include the ability to adjust discrete positions or angles.

[0086] The support plate 206 allows the sample 204 having a coated surface to be placed on the support plate 206, so that the coated surface can be positioned in a predetermined location relative to the camera 202 and / or one or more light sources 208. When the opening into which the sample 204 is inserted is closed, the housing of the apparatus forms a device that protects against the intrusion of ambient light.

[0087] Figure 3 shows a device 300 for coating a coating surface 204 according to another embodiment. A sealed space is formed by the device housing 218, which has an opening 302 that is not closed during photographic acquisition. Preferably, the opening 302 is located on the side 202 of the housing, and in the example shown, the bottom side is configured to contact the coating surface of a sample or other substrate, such as the surface of a coated workpiece. Thus, the surface 212, also referred to as the contact surface, can be formed at least partially as a flat surface, so that the contact surface 212 forms a planar contact with the coating surface 204 and substantially prevents the intrusion of ambient light passing through the opening 302. For example, the device 300 shown in Figure 3 can be placed on a horizontal surface, which has the advantage that the device can be stably located in that position without requiring further fixing means for taking photographs.

[0088] However, it is also possible to rotate the device by 90° so that the opening 302 is oriented horizontally, and then press the device against the coating surface, for example, to bring it into contact with a vertical surface.

[0089] Preferably, the device is adapted to stably and reversibly mount the device 300 to a vertically oriented surface for at least a period of time until a photograph is taken, without damaging the coated surface. In one embodiment, fastening means such as hooks, suction cups, or Velcro® fasteners can be attached to the device. For example, the device may be provided with suction cups which can be attached to a smooth surface such as a coated car door or window glass.

[0090] In some embodiments, the opening 302 is provided with a closure device, such as a door. The closure device allows for the transport of coating samples smaller than the housing of the apparatus, and, if necessary, the transport of the sample carrier into the apparatus. In this case, the closure device ensures that light does not pass through the opening 302. Thus, such an apparatus can be used to bring a larger coating surface into contact with the apparatus and to take a photograph of the coating surface through the opening 302, and to take a photograph of the coating surface of a small sample transported into the apparatus. For example, the sample transported into the apparatus may be placed on the lift 210 as shown in Figure 2, or on the bottom side 212 of the apparatus. This dual-use apparatus has the advantage of being usable for both inspection of coating defects on small samples and inspection of coating defects on larger workpieces, while maintaining the formation and optimization of the coating formulation agent, thereby allowing for the acquisition of photographs under operating conditions / modes that are optimal, or at least suitable, for the inspection of one or more types of coating defects of interest.

[0091] Figure 4 shows an apparatus for coating the coating surface according to a further embodiment. The sealed space is formed by the combination of the housing 218 of the apparatus 400 and the portable communication device 403 positioned above the camera housing opening 404, and in principle, ambient light cannot penetrate into the inside of the housing.

[0092] The apparatus 400 shown in Figure 4 may be substantially the same as, for example, the apparatus shown in Figure 2. In addition to, or separately from, the in-apparatus camera 204, the apparatus 400 has a camera housing opening 404. The camera housing opening is permanently open or closable within the wall of the housing 218 of the apparatus 400. Preferably, the wall containing the camera housing opening 404 is the wall opposite the coated surface 204, for example, the ceiling 214 of the apparatus.

[0093] Preferably, the communication device 403 is oriented and positioned such that when the communication device 403 is placed on the outer surface of the wall 214, its camera 402 can take photographs through the camera housing opening 404.

[0094] A document camera can similarly be used as an external camera instead of the camera of the communication device. Preferably, the document camera has a camera arm which is height adjustable and / or can change the orientation of the camera. Preferably, the camera housing opening 404 of the device is smaller than the component of the document camera that houses the camera lens, and the camera lens of the document camera is positioned directly above the camera housing opening, so that the camera housing opening is substantially completely covered when a photograph of the inside of the device is taken. Preferably, the document camera supports a predetermined, separate set of positions and / or orientations of the document camera, and in particular supports a portion of the document camera including the camera lens.

[0095] In some embodiments, the device 400 may have fastening means such as notches, steps, rails, Velcro, buckles, and magnets, which are configured to hold the communication device 403 in place and / or to facilitate proper positioning of the communication device by the user above the opening 404, and the camera 402 is guided toward the interior of the device 400 to take an image of the coated surface 204.

[0096] In another embodiment, any combination of the individual components of the apparatus shown in Figures 2, 3, and 4 is possible. For example, the apparatus shown in Figure 4 may have a coating surface opening 302 as shown in Figure 3 instead of the lift 210. One or more light sources may be present in a permanent / unchanging or variable arrangement or orientation. In addition to the option of positioning the camera 402 of the communication device 402 above the camera housing opening 404, or instead, the apparatus 400 may include one or more integrated cameras (not shown). These cameras can capture images of the coating surface 204 at specific image acquisition angles that cannot be achieved, for example, by the camera 402 of the communication device 403.

[0097] Figures 5A and 5B show photographs of the same coated surface, respectively. The two photographs were taken under different environmental parameters. Comparing photograph A (taken using a flat lighting angle of approximately 20° to 50°) with photograph B (taken using a steep lighting angle of approximately 80° to 110°), it can be seen that the visibility of the illustrated defects, in this case bubbles, depends on the photographic conditions.

[0098] Figure 6 shows two photographs of the same surface, one of which includes labels indicating surface defects. Photograph 600 shows a photograph of the coated surface before manual annotation (label assignment) of defects. Digital image 600 shows the coated surface with bubble defects. The coated surface contains two major defects (pores) and several smaller defects (small pores). In addition to defects, the digital image contains artifacts 608, which may be caused by one of the lenses of the image acquisition system or by dust speckles on the substrate.

[0099] When the substrate is illuminated from the side, defects become clearly visible due to shadow formation. Shadow formation allows for the identification and distinction of the coating height and depressions on the substrate. Furthermore, in the case of depressions, for example, shadow formation can be used to determine whether there are sharp edges or whether the coating thickness is gradually decreasing. This allows for the differentiation of bubble defects from crater defects.

[0100] Figure 602 shows Figure 600 in which bubble defects are manually labeled (annotated) by circles surrounding each defect. The step of annotating a digital image may include a step of storing the digital image in association with information representing the location and type of coating defects shown in the image. Where necessary, the labels may include additional data such as the mode of operation of the apparatus used to obtain the image and / or relevant parameters such as image resolution, light intensity, and illumination angle, which allow for the determination of quantitative defect measures such as diameter and circumference. For example, defects may be marked using the software VIA-VGGImage Annotator (Abhishek Dutta and Andrew Zisserman, 2019, “The VIA annotation software for image, audio and video”, Proceedings of the 27th ACM International Conference on Multimedia (MMACM, New York, NY, USA, p. 4, https: / / doi.org / 10.1145 / 3343031.3350535).

[0101] Figure 7 shows an apparatus 700 for coating a coating surface 204 according to one embodiment. For example, the apparatus 700 may be configured as described in the embodiments or examples of the invention relating to the apparatus shown in Figures 2 to 4. For example, the apparatus 700 may be operably coupled to a controller module 702. For example, the controller module 702 may be implemented as a data processing system, for example, a computer such as a single-board computer, where the data processing system is the integrated part of the apparatus 700. However, in another embodiment, the controller module may be part of an external data processing system that is communicably connected to the apparatus 700, for example, as shown in Figure 8. For example, the controller module 702 may have a configuration 704 in the form of one or more configuration files or other data structures. The configuration 704 includes a plurality of assignments. Each assignment assigns the apparatus configuration to one of a plurality of predetermined coating defect types. Each device configuration has one or more parameter values, which specify the state of one or more device components, such as a camera, light source, or lift height. These states may include, for example, the position, orientation, height, and / or light intensity of the device components, the optical or physical zoom of the camera, or the type and number of spectral filters specified for individual cameras. Setting the operating mode of the device 700 to a particular mode may involve modifying the position, orientation, or other characteristics of the device components, the values ​​of which match the values ​​or ranges of values ​​specified in the configuration for a particular defect type.

[0102] In one embodiment shown in Figure 7, the controller module 702 is configured to form a graphical user interface 706. For example, the GUI 706 may be configured such that the controller module 702 first accesses the configuration 704 to identify a list of supported defect types. For each of the supported defect types, a selectable GUI element, such as a checkbox, is generated. Each selectable GUI element allows the user to select one or more defect types of particular interest to the user. The controller module is configured to receive the user's selection and uses the configuration 704 to read out one or more device configurations associated with each of the selected defect types. Next, for each of the selected defect types, the controller module 702 causes the device 700 and its components to enter the operating mode specified in the device configuration for that defect type. This mode of operation means that the device 700 forms a photographic environment suitable for taking photographs that enable accurate detection and / or quantification of the corresponding defect type.

[0103] Furthermore, in some embodiments, the controller module 702 has or is operably coupled to image analysis software 708. This may have the advantage of not relying on network connectivity to available external devices or servers. In a building or part of a building, cellular connectivity or other reliable communication links may be unavailable for security reasons or due to the nature of the building's structure. Also, using cellular connectivity to transmit captured images to image analysis software over the network and receive results can cause significant delays. Therefore, in application scenarios where reliable and / or rapid output of analysis results regarding the identification and / or quantification of defects present on a coating surface is desired, a fully integrated solution, such as that shown in Figure 7, may be significant. For example, the image analysis software 708 may analyze one or more photographs provided as digital images to determine the presence and / or quantity of a given defect type. Quantifying defects may include determining the percentage of captured photographs affected by the defects and / or determining the number of specific individual defects (scratch, bubble) per unit area of ​​the image region.

[0104] For example, images captured by the camera of the apparatus 700 may be stored in an internal (or, in some embodiments, external) data storage device of the apparatus. The photographs may be stored in association with the ID of the mode of operation and / or one or more apparatus parameters set when the apparatus is in a given mode of operation. The image analysis software may use the ID of the mode of operation or apparatus parameters during image analysis to determine whether a particular photograph was taken during a mode of operation that enables reliable detection and / or quantification of a particular defect type. If such detection and / or quantification of a particular defect type is not possible in the mode of operation set during photograph acquisition, the photograph in question is not used as a criterion for detecting that defect type. Preferably, if the user selects multiple defect types, each requiring multiple different modes of operation, multiple photographs of the same coating surface are taken, and for each defect type of interest to the user, there is at least one photograph that can be provided as input to the image analysis software.

[0105] In one embodiment, at least one photograph is taken for each supported mode of operation of the device 700. This has the advantage that suitable photographs exist for all supported coating defects and can be used as input for analysis. However, in another embodiment, only these operating modes are set as pre-selected by the user (using the controller module 702 manually or automatically). This may have the advantage that only defect types of actual interest to the user are triggered by the device and enter a specific operating mode. This can significantly reduce the time required to capture images for analysis and save storage space.

[0106] In one embodiment, the controller module 702 has an export interface for exporting photographs taken by one or more cameras 202, 204.

[0107] Figure 8 shows a system comprising a device 800 and a portable communication device 802. For example, the portable communication device 802 may be a notebook computer, a tablet computer, or a mobile phone, particularly a smartphone. The controller module 702 may be installed on the communication device, for example, as a software application, and may substantially have the functions described with respect to Figure 7. For example, the communication device and / or the controller module 7 installed thereon may be configured to form a graphical user interface 706 through which a user may select one or more defect types, and / or through which the user may output the results of an automated inspection of images captured by the device 800 using image analysis algorithms specific to one or more defect types. In some embodiments, the controller module has image analysis software 708 having image analysis algorithms specific to one or more defect types. However, the image analysis software may not be installed on the communication device 802 but on another computer system that is communicably connected to the communication device 802.

[0108] In some embodiments, the communication device 802 has a configuration 704 and a controller module 702, the controller module configured to use the configuration to set a mode of operation for the device 800, or to switch between different modes of operation depending on the type of defect the user wishes to identify. For example, the device 800 may have a data processing system, such as a computer, which may be a single-board computer running software or firmware 804 that is interoperable with the controller module 702. For example, the interface 804 may be used to transmit photographs taken by the device 800 in the form of digital images to the controller module 702 and / or to receive control commands from the controller module 702 to set a mode of operation for the device 800.

[0109] For example, interface 804 and communication device 802 may be formed via an interface to near-field communication such as Bluetooth®. However, interface 802 and communication device 804 can also communicate via an internet connection such as cable coupling and / or WLAN-based connection.

[0110] For example, the device may support the following operating modes and their respective device settings, and may yield good detection results for crater defects, microbubble defects, and macrobubble defects. The controller module may also control switching between different operating modes of the device, or generate outputs that guide the user to change the operating mode of the device or one or more of its components. A: Detection of craters and inconsistencies: Place the coated glass plate on a light table positioned on top of the lift. Adjust the camera position and / or lift height so that the distance between the camera and the sample is approximately 7.5 cm. Take one or more photographs. Craters and inconsistencies are detected as bright areas in the photographs. B: Microbubble detection: The coated sample is illuminated from the side at a flat angle, and the position of one or more light sources and / or the placement of the sample and / or the camera are adjusted so that the distance between the camera and the sample is approximately 7.5 cm. One or more photographs are taken. Bubbles are detected by the formation of shadows due to oblique incidence of light. C: Macrobubble detection: The settings are the same as for microbubble detection, but the distance between the camera and the sample is set to approximately 17 cm. The difference in distance between microbubble detection and macrobubble detection ensures that microbubbles can be reliably visualized with sufficiently high resolution in the photograph, and also ensures that a large area of ​​the sample is reliably analyzed during macrobubble detection in order to detect any separated bubbles.

[0111] The aforementioned distance specifications may be modified depending on the camera and optical system included in each device.

[0112] Figure 9 shows a system comprising a server computer 902, multiple devices 900, 908, and multiple communication devices 802, 906. The system shown in Figure 9 comprises the system components described with reference to Figure 8, plus an additional server computer 902 and other system components 906, 908. Preferably, the image analysis software 708 is demonstrated on the server computer 902 rather than on the device 900 or communication device 802. This is because the server computer typically has greater computing power than the device 900 or smartphone. The controller module 702 on the communication device is interoperable with interface 804, receiving digital images of the surface coating from the interface and / or sending control commands to interface 804 to determine the mode of operation of the device 900. Furthermore, the controller module 702 is interoperable with the server computer 902, which transmits the digital images of the coating surface received from the device 900 to the server computer 902 via the network, where the server computer performs image analysis. The analysis results are sent back to the controller module 702 via the network by the image analysis software 708 on the server 902. The controller module then outputs the results directly to the user via the GUI 706 of the communication device 802, and / or sends the results to the interface 804 of the device 900, which can then output the results to the user via the display of the device 900.

[0113] Typically, the system may have multiple other devices 908 and multiple other portable communication devices 906, each of which is communicatively coupled to a corresponding device, and image data, control commands, and / or image analysis results are exchanged with the corresponding device.

[0114] In some embodiments, the apparatus 900, the portable communication device 802, and / or the server computer 902 may also include a program and software framework for performing a machine learning process ("machine learning framework 904"). For example, the machine learning framework 904 may include software that allows a user or an automated user to run the program to detect and annotate coating defects in digital images. Photograph 602 in Figure 6 shows a digital image of surface defects that has already been annotated. The framework is configured to receive digital images generated by one or more apparatuses 900, 908. For example, a neural network may be trained on annotated images of various coating defect types, and a predictive model created during training may be able to identify and / or quantify corresponding coating defects in input images. The predictive model may be used, if necessary, to be integrated into a software application along with other models specifically trained to detect and characterize other coating defect types, which may then be used as image analysis software 708.

[0115] In some examples, the apparatus shown in Figures 2 to 4 and Figures 7 to 9 can be used to generate training data sets. Using the described apparatus, digital images of many different coating surfaces containing many different types of coating defects are acquired. For bubble defects, a flat angle of incidence of light may be selected. For other types of defects, other image acquisition settings and conditions may be selected. Preferably, a large number of digital images showing thousands of defects of different defect types and different types of coating substrates are acquired and manually annotated (labeled).

[0116] Figure 10 shows a GUI 706 in which the user can select one or more defect types. For example, the GUI 706 may be configured in the first window 950 to display to the user a list of defect types when the apparatus shown in Figures 2 to 4 or Figures 7 to 9 has an appropriate mode of operation. This means that it is possible to assume a state in which a photograph of the coating surface can be taken on the apparatus or a component contained within the apparatus under conditions in which a specific defect type can be detected with sufficient accuracy. Here, multiple defect types can be detected using a certain mode of operation.

[0117] In the GUI shown in Figure 10, a user-selectable GUI element, such as a checkbox, may be displayed for each defect type supported by the device. In some examples, after the user selects multiple defect types, the controller module sequentially assumes an operating mode and / or configuration of the device components that is suitable for taking photographs (digital images) of each of the selected defect types that can be identified by image analysis software, either automatically or semi-automatically. Modifications to the configuration or operating mode of the device components may correspond to and represent different operating modes of the device. For example, after the user selects three defects, "macrobubble defects," "microbubble defects," and "cratering defects," the controller module may first configure the device in a first operating mode suitable for taking photographs of microbubble defects that are easily visible and characterizable. In the next step, after the device has taken and stored or transferred at least one photograph in the microbubble defect operating mode, the controller module switches the device to another operating mode suitable for taking photographs of macrobubble defects that are easily identifiable. After at least one photograph has been taken in this operating mode, the controller module ensures that the apparatus switches to another operating mode suitable for taking photographs in which crater defects can be easily identified. In some embodiments, at least one of the assumed arrangements of apparatus members during which the controller triggers a sequence of changes in the apparatus's operating mode is a specific height of the sample carrier for the defect type. After photographs of the selected defect type have been taken by the apparatus, they are analyzed by image analysis software 708, which may be installed in the apparatus or an external data processing system 802, 902, and the presence and / or severity of the defect is determined for each of the selected defect types in the images taken, corresponding to the type of impact. The results of this analysis may be output via GUI 706 to, for example, another window 952.

Claims

1. A method for inspecting a coated surface for surface defects, - A step of using a device that covers a coated surface to be inspected, wherein the device forms a closed space, isolates the coated surface to be inspected from ambient light, provides predetermined photographic acquisition conditions within the closed space, the device has a controller module and / or is operably coupled to the controller module, the controller module is configured to control the operation of the device, - A step of providing a configuration having multiple assignments, wherein each assignment assigns one of a plurality of different coating defect types to one of a plurality of device settings, and each device setting determines one or more device configuration parameters that specify the position and / or mode of operation of one or more components of the device, - The controller module receives a selection of at least one coating defect type, - The controller module identifies the device settings stored in the configuration in association with the selected coating defect type, - A step of changing at least one component of the apparatus by the controller module, wherein the at least one component is selected from a group including a camera, a light source, and / or a sample carrier, according to the identified apparatus setting, thereby setting the mode of operation of the apparatus to defect type identification image acquisition mode. - A step of acquiring a photograph of the coating surface located within the enclosed space, wherein the step of acquiring the photograph is performed while the apparatus is in defect type identification image acquisition mode, - A step of inspecting the photograph to determine whether or not there are surface defects, It has, The apparatus has a sample carrier configured to transport a sample having a coated surface to be inspected. The sample carrier is a manually adjustable lift, and the lift has manually movable rotating wheels that allow for height adjustment.

2. - The apparatus has a housing, the housing is opaque to light, the steps of using the apparatus include the step of covering the coating surface, and / or - The apparatus has one or more light sources, and / or The method according to claim 1, wherein the apparatus has a camera housing opening that allows images of the coating surface to be taken with at least one in-apparatus camera and / or at least one out-apparatus camera, and the step of taking the photograph is performed with the at least one camera.

3. - The apparatus has a sample carrier for transporting a coating sample including the coating surface, and the step of using the apparatus to cover the coating surface includes the step of placing the sample on the sample carrier, and / or - The apparatus has a coating surface opening, the coating surface opening is an opening in the wall of the housing referred to as a contact wall, the contact wall is intended to be in contact with the coating surface, and the step of using the apparatus covering the coating surface includes the step of bringing the contact wall into contact with the coating surface, and / or The method according to claim 2, wherein the apparatus has a sample position marking, the sample position marking indicates a position within the apparatus where the sample having the coated surface is placed.

4. moreover, - A step of outputting a message to the user via the controller module, the message representing a method for changing the position, orientation, and / or mode of operation of at least one component of the apparatus according to the identified apparatus settings, wherein the at least one component is selected from a group including a camera, a light source, and / or a sample carrier, and the message enables the user to manually or semi-automatically set the mode of operation of the apparatus to a defect type identification image acquisition mode. It has, The method according to claim 1, wherein the acquisition of the aforementioned photograph is performed while the apparatus is in the defect type identification image acquisition mode.

5. The aforementioned device configuration parameters are: - Identification of one or more of the multiple light sources included in the device that are turned on, - Identification of one or more of the multiple light sources included in the device that are to be turned off, - The light intensity of one or more of the multiple light sources included in the apparatus, - The position of each of one or more light sources of the plurality of light sources included in the apparatus, in particular the height and / or horizontal offset of each light source with respect to one or more walls of the housing, - The distance between each of the one or more light sources of the plurality of light sources included in the apparatus and the coating surface, - The irradiation angle between each of the one or more light sources of the plurality of light sources included in the apparatus and the coating surface, - Position and / or height and / or inclination of the sample carrier within the apparatus, - Identification of one or more in-device cameras included in the apparatus described in claim 2, wherein the in-device camera is used for image acquisition, identification, - Identification of one or more in-device cameras included in the apparatus described in claim 2, wherein the in-device cameras are not used for image acquisition, identification, - The arrangement of one or more of the multiple in-device cameras included in the apparatus according to claim 2, in particular the height and / or horizontal offset of each camera with respect to one or more of the walls of the housing, - The distance of one or more of the multiple cameras included in the apparatus to the coating surface, - The image acquisition angle between each of the multiple cameras included in the apparatus and the coating surface, The method according to claim 2, selected from the group including the following.

6. The method according to claim 1, wherein the device has a display configured to display a GUI formed by the controller module.

7. The method according to claim 1, wherein the apparatus has a portable communication device and / or is operably coupled to the portable communication device, and the portable communication device has the controller module.

8. Furthermore, the controller module is configured to form a graphical user interface GUI, wherein the GUI is configured to display a plurality of selectable GUI elements, each representing one of the plurality of different coating defect types. It has, The method according to claim 1, wherein the controller module receives a selection of at least one type of coating defect in the form of a user's selection of each of the GUI elements.

9. Furthermore, in response to the user selecting multiple defect types via the GUI, the controller module causes the component to automatically and sequentially assume a mode of operation and / or position suitable for capturing images in which each of the selected defect types is identified by the image analysis software. After at least one photograph has been taken in a predetermined operating mode of the apparatus, the controller module ensures that the apparatus switches to another operating mode suitable for taking a photograph in which the next selected type of defect can be identified. In particular, the method according to claim 8, wherein at least one of the assumed positions of the member is a height specific to the defect type of the sample carrier.

10. moreover, - A step of acquiring a digital preview image using at least one camera, wherein the preview image shows the coating surface, - A step of performing an initial image analysis of the preview image using the controller module, wherein at least one type of coating defect shown in the preview image is identified. It has, The method according to claim 1, wherein the at least one selected coating defect type is at least one coating defect type identified in the initial image analysis.

11. The inspection of the aforementioned photograph is performed using image analysis software and includes the step of obtaining qualitative and / or quantitative characteristics of the coating defect type on the coating surface. The method according to claim 1, further comprising the step of outputting the results of image analysis using the image analysis software.

12. The rotating wheel has separate markings indicating a predetermined number of different arrangements of the lifts and corresponding heights, The method according to claim 1, further comprising the steps of: the controller module determining, in response to a user selecting a particular defect type, that the lift has a predetermined position specifically assigned to the selected defect type; and the controller module informing the user via a GUI that the rotating wheel of the sample carrier needs to be rotated until one of the marks on the rotating wheel representing the determined position reaches the target position.

13. The device has multiple light sources, One of the aforementioned multiple light sources is a light panel, The light panel is a sample support plate of a sample carrier, or the light panel is placed on the sample carrier. The method according to claim 1, wherein the light panel functions as a transmitted light source that emits light passing through the coating surface before it is acquired by a camera to form the photograph.

14. The method according to claim 1, wherein the apparatus has a plurality of light sources, the light sources having different arrangements and / or orientations, so that one or more of the light sources can be individually controlled with respect to a turn-on state, a turn-off state, and / or light intensity.

15. The apparatus is used to acquire multiple photographs of multiple coated surfaces. This method further, - A step of storing the acquired photograph as a digital image associated with an image label in a storage medium, wherein the image label represents the location and type of one or more surface defects of one or more different coating defect types, A step comprising inputting the acquired digital images and their associated labels into a machine learning program, wherein the machine learning program is configured to perform a machine learning method that learns the relationships between the image features of the acquired digital images and the coating defect types and / or quantities indicated in the labels, and the learned relationships are stored as a trained predictive model, A step of integrating the predictive model into image analysis software, wherein the image analysis software is configured to qualitatively and / or quantitatively characterize defects on the coating surface of a sample shown in a digital image provided as input; The method according to claim 1, having the following characteristics.

16. It is a system, The apparatus has a configuration that forms a closed space that isolates the coated surface from ambient light and provides predetermined photographic acquisition conditions within the closed space. The aforementioned device is A housing that is opaque to light, One or more light sources, The housing comprises at least one in-device camera and / or camera housing opening, wherein the camera housing opening allows at least one external camera to acquire an image of the coating surface through the camera housing opening. A sample carrier for transporting the coating sample, and sample position marking, A controller module configured to control the operation of the aforementioned device, A configuration comprising multiple assignments, each assignment assigning one of several different coating defect types to one of several device settings, and each device setting determining one or more device configuration parameters that specify the position and / or mode of operation of one or more components of the device, It has, The sample carrier is a manually adjustable lift, and the lift has manually movable rotating wheels that allow for height adjustment. The aforementioned sample position marking indicates the position within the apparatus where the sample having the coated surface is placed. The controller module is Upon receiving at least one selection of the aforementioned coating defect type, Identify the device settings stored in the configuration in relation to the selected coating defect type, In accordance with the identified device settings, at least one component of the device, selected from the group including a camera, light source, and / or sample carrier, is modified so that the operating mode of the device is set to defect type identification image acquisition mode. It is configured in such a way, The system is configured such that image acquisition is performed while the device is in the defect type identification image acquisition mode.

17. moreover, The controller module has a computer system operably coupled to it, The computer system includes image analysis software, which is configured to perform image analysis to identify and / or quantify one or more coating defect types on the coating surface shown in the image. The system according to claim 16, wherein the computer system is in particular a server computer or a portable communication device.

Images