Inspection system for inspecting a lighting device during an assembly process of the lighting device and method therefor

Abstract

An inspection system (100) for inspecting a lighting device (150) during an assembly process of the lighting device; wherein the lighting device comprises a substrate (153) and a plurality of components (155) mounted on the substrate; wherein the inspection system comprises: a light source (120) arranged for illuminating the lighting device according to a first light output spectrum to provide a luminance contrast between the substrate and the plurality of components; an imaging unit / camera (130) arranged for capturing a first image of the illuminated lighting device; a controller (110) comprising a processing unit for determining a luminance contrast measure of the captured first image; wherein the processing unit is further arranged for adjusting the first light output spectrum when the luminance contrast measure of the captured first image exceeds a threshold value, and wherein the imaging unit is further arranged for capturing a second image of the lighting device illuminated according to the adjusted first light output spectrum; and wherein the controller further comprises: a comparison unit arranged for comparing the second image with a reference image; a determination unit arranged for determining a defect in the substrate and / or in the plurality of components based on the comparison.

Description

Technical Field

[0001] This invention relates to an inspection system for inspecting lighting equipment during the assembly process. The invention further relates to a method and computer program product for inspecting lighting equipment during the assembly process. Background Technology

[0002] LEDs (Light Emitting Diodes) are solid-state semiconductor devices that convert electrical energy into visible light. Due to their small size, low power consumption, long lifespan, high brightness, low heat generation, environmental friendliness, and durability, LEDs have become an ideal light source to replace traditional light sources. LEDs are used quite flexibly and can be manufactured into small, thin products in various forms, including points, lines, and surfaces. LEDs are widely used in various types of lighting equipment, such as battery-powered flashlights, small voice-activated lights, safety flashlights, street and indoor stair lighting, and continuous lighting for buildings and landmarks.

[0003] As demand for LEDs (or LED-based lighting devices) increases, so does their production. To automate end-of-line quality control and process execution in LED (or LED-based lighting device) production—for example, checking the execution of required process steps and the quality of those steps—computer vision (or machine learning) technologies are being applied. Visible light-based imaging techniques are commonly used in such technologies.

[0004] US 2004 / 184031A1 discloses a three-dimensional optical inspection system that reconstructs a three-dimensional image of the surface shape of an object, at least partially specularly reflective, located on a printed circuit board, by capturing two or more two-dimensional images of the object under different lighting configurations. Diffuse reflection and specular reflection can be used to reconstruct the three-dimensional image using any reconstruction method, such as photometric stereo. The different lighting configurations can be implemented using a light source comprising light-emitting elements arranged in a concentric circular array, wherein each circular array is divided into multiple sections. Each section is independently controlled to selectively activate these sections, thereby illuminating the object in a pre-established lighting pattern. Summary of the Invention

[0005] The inventors have recognized that the use of computer vision technology in automating the LED assembly process has significant limitations in identifying defects in LEDs (or LED-based lighting devices), leading to, for example, false defect detection or failure to detect defects. For instance, this becomes extremely difficult when it is necessary to identify and capture features of white objects (e.g., white-painted electronic drivers) against a white background (a white-coated metal substrate). Or, as another example, identifying a quasi-transparent lens plate on top of a printed circuit board and analyzing, for example, the alignment of the lens plate with features on the printed circuit board is very difficult.

[0006] Therefore, the object of the present invention is to provide a system having improved identification of defects in (LED-based) lighting devices during the assembly process, for example for automatically tracking the assembly process of lighting devices.

[0007] According to a first aspect, this objective is achieved by an inspection system for inspecting a lighting device during the assembly process of the lighting device; wherein the lighting device includes a substrate and a plurality of components mounted on the substrate. The inspection system includes: a light source arranged to illuminate the lighting device according to a first light output spectrum to provide brightness contrast between the substrate and the plurality of components; an imaging unit arranged to capture a first image of the illuminated lighting device; and a controller including a processing unit for determining a brightness contrast metric of the captured first image. The processing unit is further arranged to adjust the first light output spectrum when the brightness contrast metric of the captured first image exceeds a threshold, and the imaging unit is further arranged to capture a second image of the lighting device illuminated according to the adjusted first light output spectrum. The controller further includes: a comparison unit arranged to compare the second image with a reference image; and a determination unit arranged to determine defects in the substrate and / or in the plurality of components based on the comparison.

[0008] The inspection system includes a light source for illuminating a lighting device according to a first light output spectrum. The light source may be external to the lighting device. The first light output spectrum may include a first wavelength or a first wavelength range. For example, the first light output spectrum may be blue with wavelengths between approximately 380 nm and 500 nm. The first light output spectrum can be used to provide brightness contrast between a substrate and multiple components. Brightness contrast is the difference in brightness or color that makes an object (or its representation in an image or display, such as a substrate and multiple components in an image) distinguishable.

[0009] The inspection system further includes an imaging unit arranged to capture a first image of the illuminated lighting equipment. The imaging unit may be a camera. It may be a 2D video camera, a stereo video camera, or a depth-sensing (range-finding) video camera (e.g., a time-of-flight camera). The imaging unit may include one or more imaging devices. The illuminated lighting equipment may be illuminated using a first light output spectrum.

[0010] The inspection system further includes a controller comprising a processing unit for determining a luminance-contrast metric of the captured first image. The luminance-contrast metric may include pixel intensities of the image, such as the pixel intensities of the substrate of a lighting device and multiple components. A pixel is a physical point in an image, or the smallest addressable element in a display device where all points are addressable; therefore, it is the smallest controllable element of the image represented on the screen. For example, a pixel in a grayscale image may require one byte, for instance, to represent the amount of grayscale intensity in order to render the pixel on the screen. Pixels in a color image are represented by three values: red, green, and blue (r, g, b). These values ​​indicate the intensity of red, green, and blue required to render the pixel on the screen, respectively. The luminance-contrast metric may include (sudden) changes in pixel intensity in the image, such as changes in pixel intensity of multiple components relative to the substrate. Additionally and / or alternatively, the luminance-contrast metric may include one or more of the following: Weber contrast, Michelson contrast, RMS contrast, etc.

[0011] The processing unit may be further arranged to adjust the first light output spectrum, thereby optimizing the first light output spectrum, when the brightness contrast metric of the captured first image exceeds a threshold. The imaging unit may (then) be further arranged to capture a second image of the lighting device illuminated according to the adjusted first light output spectrum. The controller may further include: a comparison unit, which may be arranged to compare the second image with a reference image; and a determination unit, which may be arranged to determine defects in the substrate and / or multiple components based on the comparison. Because the inspection system optimizes the light output spectrum of the light source and determines defects in the substrate and / or multiple components based on the captured image of the lighting device according to the optimized light output, an improved system is provided that has improved identification of defects in (LED-based) lighting devices during the assembly process. The provided system is not limited to LED-based lighting devices but also provides improved defect identification for other lighting devices.

[0012] In an embodiment, the comparison unit may be arranged to compare the first image with a reference image when the brightness contrast metric of the captured first image does not exceed a threshold.

[0013] In this example, the brightness and contrast metric of the captured first image may be sufficient to inspect the substrate and / or multiple components. In other words, the first light output spectrum may be optimal for the substrate and / or multiple components and therefore does not require adjustment. For example, the first light output spectrum may be based on expert knowledge. Therefore, in this case, anomaly detection may be based on a comparison of the first image of the lighting device with a reference image. The reference image may include an image of the lighting device that is free of defects in the substrate and multiple components.

[0014] In an embodiment, the processing unit may be arranged to obtain signals indicating the optical properties of the substrate and / or multiple components; and to adjust the first light output spectrum based on the obtained optical properties.

[0015] Signals indicating the optical properties of the substrate and / or components can be obtained. The optical properties of a material define how it interacts with light. For each material—such as the substrate and / or components—incident radiation is partially transmitted, partially reflected, and partially absorbed. Therefore, optical properties can include transmittance, reflectance, and absorptivity, among others. Based on these optical properties, the first light output spectrum can be advantageously adjusted to improve brightness contrast.

[0016] In an embodiment, the processing unit may be arranged to adjust the first optical output spectrum by sequentially increasing or decreasing it. Alternatively or additionally, to adjust the first optical output spectrum based on the obtained optical characteristics, the processing unit may sequentially scan all spectra (wavelengths) and use a hit-and-trial method to optimize the first optical output spectrum. The spectral scanning may be performed via a feedback loop.

[0017] In this embodiment, the lighting equipment may be de-energized during the assembly process.

[0018] During the assembly process, the lighting device does not receive electricity; for example, it is not powered. Multiple components mounted on the substrate (such as LEDs) can illuminate the lighting device without being powered. Therefore, an external light source can be used to illuminate the lighting device, and the initial light output spectrum can be optimized to improve brightness contrast.

[0019] In one embodiment, the multiple components may include at least one phosphor-coated LED.

[0020] Multiple components may include at least one phosphor-coated LED. In other examples, multiple components may further include (multiple) electronic LED drivers, screwdrivers, wires, etc. The LED may be coated with phosphor, for example, to convert blue into a mixed white. While yellow phosphors provide some contrast between the LED and the substrate, they are inefficient.

[0021] In an embodiment, the first light output spectrum may be a blue light spectrum or a UV light spectrum.

[0022] In the example, the first light output spectrum could be a spectrum that causes multiple components to emit light that can be detected by the imaging unit. For example, for a phosphor-coated LED, using the blue light spectrum can provide a detectable image with improved brightness contrast for anomaly detection. In another example, the UV spectrum could be used to improve the brightness contrast of fluorescent components (such as labels, paints, and adhesives) among multiple components.

[0023] In an embodiment, the comparison unit may be arranged to perform image comparison using computer vision and / or machine learning algorithms.

[0024] Various state-of-the-art computer vision and / or machine learning algorithms can be used for image comparison. For example, generative adversarial networks (GANs) can be used as machine learning algorithms to identify anomalies (defects) in a substrate and / or multiple components.

[0025] In this embodiment, the threshold may be determined by computer vision and / or machine learning algorithms.

[0026] In this advantageous embodiment, the threshold for the brightness contrast measurement can be determined by computer vision and / or machine learning algorithms. For example, the threshold can be determined when a specific level of accuracy is achieved by the computer vision and / or machine learning algorithm. Alternatively, the threshold can be selected by the user.

[0027] In an embodiment, the light source may be a multispectral light source, which is arranged to illuminate the lighting device at least according to a first light output spectrum and a second spectrum; wherein, the processing unit may be further arranged to adjust the first light output spectrum and / or the second spectrum when the brightness contrast metric of the image of the lighting device illuminated by the multispectral light source exceeds a threshold.

[0028] In the example, the light source can be a multispectral light source. A multispectral light source is a light source capable of emitting more than one spectral component of light. In an alternative example, the light source may include multiple light sources arranged to emit multiple spectra. In the example, a first light output spectrum can provide background illumination to make all the multiple components visible, and a second light output spectrum can provide increased brightness contrast for one or more of the multiple components (indistinguishable by means of the first light output spectrum). In the example, the first light output spectrum can be white light, and the second light output spectrum can be a UV spectrum, for example, to enhance phosphorescence. In another example, the multispectral light source can be arranged to use polarized light.

[0029] In this embodiment, the imaging unit may be a multispectral sensing device.

[0030] Multispectral images are images that capture image data across a specific wavelength range of the electromagnetic spectrum. Multispectral sensing devices can be advantageously used to capture multispectral images of lighting equipment.

[0031] In one embodiment, the imaging unit may include a thermal imaging device arranged to capture a first thermal image of the illuminated lighting device illuminated according to a first light output spectrum and / or a second thermal image of the illuminated lighting device illuminated according to an adjusted first light output spectrum. In another embodiment, a comparison unit may be further arranged to compare the first or second thermal image with a reference thermal image; and a determining unit may be further arranged to determine defects in the substrate and / or in multiple components based on the comparison.

[0032] Based on the Stokes displacement of the phosphor during light conversion, photoexcitation on the lighting device can cause a temperature rise in at least several components (e.g., parts containing the phosphor). The imaging unit may include a thermal vision device (e.g., a thermal camera) to capture first and / or second thermal images of the lighting device. A first thermal image of the lighting device can be captured when the lighting device is illuminated according to a first light output spectrum, and a second thermal image can be captured when the lighting device is illuminated according to a modified first light output spectrum. These thermal images can be compared with a reference thermal image, which may represent an appropriate (defect-free) thermal distribution of the substrate and / or several components, to identify defects in the substrate and / or several components. This will provide further refinement of the determination of defects in the lighting device.

[0033] According to a second aspect, this objective is achieved by a method for inspecting a lighting device during the assembly process of the lighting device; wherein the lighting device includes a substrate and a plurality of components mounted on the substrate. The method includes the steps of: illuminating the lighting device according to a first light output spectrum to provide a brightness contrast between the substrate and the plurality of components; capturing a first image of the illuminated lighting device; and determining a brightness contrast metric of the captured first image. Furthermore, when the brightness contrast metric of the captured first image exceeds a threshold, the first light output spectrum is adjusted; and a second image of the lighting device illuminated according to the adjusted first light output spectrum is captured; wherein the method further includes: comparing the second image with a reference image; and determining defects in the substrate and / or in the plurality of components based on the comparison.

[0034] According to a third aspect, this objective is achieved by a controller for inspecting the lighting equipment during the assembly process of the lighting equipment; wherein the controller includes: an input interface and an output interface; a processing unit arranged to determine a brightness contrast metric of a first image of the lighting equipment illuminated according to a first light output spectrum; and wherein the processing unit is further arranged to adjust the first light output spectrum when the brightness contrast metric of the first image exceeds a threshold; a comparison unit arranged to compare a first or second image of the lighting equipment illuminated according to the adjusted first light output spectrum with a reference image; and a determination unit arranged to determine, based on the comparison, defects in the substrate and / or multiple components of the lighting equipment.

[0035] According to the fourth aspect, this objective is achieved by a computer program product including instructions that, when executed by a computer, cause the computer to perform the steps of the method of the second aspect.

[0036] It should be understood that computer program products and methods may have similar and / or the same embodiments and advantages as the systems described above. Attached Figure Description

[0037] Referring to the accompanying drawings, the above and additional objects, features, and advantages of the disclosed systems, devices, and methods will be better understood through the following illustrative and non-limiting detailed description of embodiments of the systems, devices, and methods, in which:

[0038] Figure 1 An embodiment of an inspection system for inspecting lighting equipment during the assembly process is illustrated schematically and exemplary.

[0039] Figure 2 An embodiment of the lighting device is illustrated schematically and exemplary.

[0040] Figure 3 An embodiment for inspecting a lighting device's controller during the assembly process is illustrated schematically and exemplary.

[0041] Figure 4 A flowchart is shown schematically and exemplary, illustrating an embodiment of a method for inspecting a lighting device during the assembly process.

[0042] All figures are schematic and not necessarily to scale, and generally only show the parts necessary to illustrate the invention, where other parts may be omitted or only suggested. Detailed Implementation

[0043] Figure 1An embodiment of an inspection system 100 for inspecting a lighting device 150 during the assembly process is illustrated schematically and exemplary. The lighting device 150 may include a substrate 153 and a plurality of components 155 mounted on the substrate 153. During the assembly process, the lighting device 150 may be without power, for example, it does not receive power. Figure 1 A side view of the lighting device 150 is shown. The substrate 153 may include a metal plate, a plastic plate, or a polymer plate. The lighting device 150 is a device or structure arranged to emit light suitable for illuminating an environment, providing or substantially contributing to illumination of a scale sufficient for that purpose. The lighting device 150 may include at least one light source or lamp, such as an LED-based lamp, a gas discharge lamp, or an incandescent bulb, optionally with any associated support, housing, or other such enclosure. For example, multiple components 155 may include LED sources, LED drivers, screws, wires, labels, adhesive materials, paint, etc.

[0044] Each lighting fixture 150 can take any of the following forms, such as ceiling-mounted fixtures, wall-mounted fixtures, wall washer lights, or freestanding fixtures (and these fixtures do not necessarily have to be of the same type).

[0045] The inspection system 100 may include multiple light sources 120 arranged to illuminate the lighting device 150 according to a first light output spectrum to provide brightness contrast between the substrate 153 and the plurality of components 155. The light sources 120 may be independent lighting devices external to the lighting device 150. The light sources 120 may include LED-based lamps, gas discharge lamps, or incandescent bulbs, etc., optionally having any associated support, housing, or other such enclosure, and may take any form. Only one light source 120 is shown in this exemplary figure. One or more light sources 120 may be used in the inspection system 100. The light source 120 may be a multispectral light source arranged to illuminate the lighting device according to a first light output spectrum and a second spectrum. Alternatively, multiple light sources may be used to emit the first light output spectrum and the second spectrum.

[0046] The inspection system 100 may further include an imaging unit 130 arranged to capture a first image of the illuminated lighting device 150 illuminated according to a first light output spectrum. The first image is a still image of the illuminated lighting device 150. The imaging unit 130 may be a camera. The imaging unit may be a 2D video camera, a stereo video camera, or a depth-sensing (range-finding) video camera (e.g., a time-of-flight camera). The use of any imaging device known in the art is not excluded. In an exemplary figure, the imaging unit 130 may include one imaging device. The imaging unit 130 may include one or more imaging devices; wherein the first image of the illuminated lighting device 150 may, for example, be a combination of multiple images from multiple imaging devices, for example, by performing synthesis. The imaging unit 130 may include a multispectral sensing device. The imaging unit 130 may include a thermal imaging device arranged to capture a first thermal image of the illuminated lighting device illuminated according to a first light output spectrum and / or a second thermal image of the illuminated lighting device illuminated according to an adjusted first light output spectrum.

[0047] The inspection system 100 may further include a controller 110. The controller 110 may include a processing unit (not shown) for determining a luminance contrast ratio of the captured first image. The luminance contrast ratio may include different measures representing the luminance contrast of the image. One or more luminance contrast ratio measures may be used. In an example, the luminance contrast ratio measure may be a pixel intensity or a change in pixel intensity. For example, the luminance contrast ratio measure may be a change in pixel intensity of a plurality of parts 155 and a substrate 153. The change in pixel intensity may be directed at one or more of the plurality of parts. The change may be a sudden or expected change in intensity. In another example, the luminance contrast ratio measure may include one or more of the following: Weber contrast ratio, Michelson contrast ratio, RMS contrast ratio, contrast sensitivity function, etc. RMS contrast ratio is related to the intensity change and is defined as the standard deviation of pixel intensity. Any other luminance contrast ratio measure known in the art may also be used.

[0048] The processing unit may be further arranged to adjust the first light output spectrum when the brightness contrast metric of the captured first image exceeds a threshold. The processing unit may be arranged to obtain signals indicating the optical characteristics of substrate 153 and / or multiple components 155. The adjustment of the first light output spectrum may be based on the obtained optical characteristics. Optical characteristics may include one or more of the following: refraction, polarization, reflection, absorption, photoluminescence (fluorescence), transmission, diffraction, dispersion, dichroism, scattering, birefringence, color, photosensitivity, etc. For example, blue light or UV light may be used for substrates / multiple components having phosphorescent or fluorescent compounds. Additionally and / or alternatively, the processing unit may be arranged to adjust the first light output spectrum by sequentially increasing or decreasing it. For example, optimization of substrate 153 / multiple components 155 can be achieved by scanning different spectra. The spectral scanning may be input into (multiple) computer vision / machine learning algorithms with the aim of defining one or more wavelengths that provide optimal feature recognition. The scanning and optimization may be referred to as automatic calibration, which may be repeated for each new lighting device or for each new change in environmental conditions (such as, for example, in another factory). In the example, different spectra can be used to identify defects in one or more different parts of the plurality of parts 155 or in the substrate 153.

[0049] Imaging unit 130 may (then) be further arranged to capture a second image of lighting device 150 illuminated according to an adjusted first light output spectrum. Controller 110 may control light source 120 to illuminate according to the adjusted first light output spectrum. Controller 110 may then trigger imaging unit 130 to capture the second image of lighting device 150.

[0050] The controller 110 may further include a comparison unit (not shown) arranged to compare the second image with a reference image. The comparison unit may be arranged to perform the image comparison using computer vision and / or machine learning algorithms. For example, the comparison unit may use pattern recognition algorithms, such as classification algorithms (e.g., linear discriminant analysis, quadratic discriminant analysis, etc.), clustering algorithms (e.g., k-means clustering, correlation clustering, etc.), generative adversarial networks (GANs), template matching, etc. The use of other computer vision and / or machine learning algorithms known in the art for detecting anomalies is not excluded. In this example, a threshold may be determined by a computer vision and / or machine learning algorithm, for example, to ensure that defect determination is performed optimally. The controller 110 further includes a determination unit (not shown) arranged to determine defects in the substrate and / or multiple components based on the comparison. Defects may include one or more of the following: missing, misaligned, misaligned, damaged, etc., of the substrate 153 and / or one or more of the multiple components 155. The determination unit may be further arranged to determine a stage of the manufacturing process by comparing the second / first captured image with multiple reference images, each reference image representing a different stage of the process. The comparison unit may be further arranged to compare the first or second thermal image with a reference thermal image; and the determination unit may be further arranged to determine defects in the substrate 153 and / or multiple components 155 based on the (thermal image) comparison.

[0051] The inspection system 100 may include one or more independent devices distributed in space, such as a light source 120, an imaging unit 130, etc. Alternatively, the inspection system 100 may be contained in a single device, such that the light source 120, the imaging unit 130, etc., are units within a single device. Furthermore, the inspection system 100 may be partially contained within a device; for example, the light source 120 and the imaging unit 130 may be contained in a single device, and the controller 110 may be contained in a device external to the light source 120 / imaging unit 130 or in a device separate from the light source 120 / imaging unit 130.

[0052] Figure 2An embodiment of a lighting device 250 is illustrated schematically and exemplary. The lighting device 250 may include a substrate 253. The lighting device 250 may further include a plurality of components 255a-h, 257a-c mounted on the substrate 253. In this exemplary figure, the plurality of components 255a-h, 257a-c includes a plurality of LED sources 255a-h and a plurality of screws 257a-c. The plurality of components 255a-h, 257a-c may further include wires (not shown), LED drivers, capacitors (not shown), resistors (not shown), etc. The interior of the lighting device 250 and its components in the optical path (e.g., substrate 253 and / or the plurality of components 255a-h, 257a-c) are typically constructed in such a manner that light efficiency is maximized. This typically means that these components are made white (unless these components will be optical components, such as mirrors (mirror metal) or shaped optics (such as clear lenses)).

[0053] Figure 3 An embodiment of a controller 310 for inspecting lighting devices 150, 250 during the assembly process of lighting devices 150, 250 is illustrated schematically and exemplary. The controller 310 may include an input interface 301. The input interface 301 may be arranged to obtain signals indicating the optical characteristics of substrates 153, 253 and / or multiple components 155, 255a-h, 257a-c. These signals may be received from an external network (such as from the cloud) or may be stored in the memory 309 of the controller 310. The controller 310 may further include an output interface 302. The output interface 302 may be arranged to output a first control signal to the light source 120 related to the emitted (adjusted) first light output spectrum, and / or output a second control signal to the imaging unit 130 related to capturing first and / or second images of the lighting devices 150, 250. The controller 310 may further include a processing unit 303 arranged to determine a brightness contrast metric of the captured first image, wherein the processing unit 303 may be further arranged to adjust a first light output spectrum when the brightness contrast metric of the captured first image exceeds a threshold. The processing unit 303 may be further arranged to control the light source 120 to illuminate the adjusted first light output spectrum. The processing unit 303 may be further arranged to control the imaging unit 130 to capture a second image of the lighting devices 150, 250 illuminated according to the adjusted first light output spectrum. Additionally and / or alternatively, the imaging unit 130 has a sensing device (not shown) to determine when the light output of the light source 120 has changed, and then captures the second image of the lighting devices 150, 250.

[0054] The controller 310 may further include a comparison unit 305, which may be arranged to compare the second image with a reference image. The comparison unit 305 may use computer vision and / or machine learning algorithms. The controller 310 may further include a determination unit 307, which is arranged to determine defects in the substrates 153, 253 and / or in the plurality of components 155, 255a-h, 257a-c based on the comparison.

[0055] The controller 310 may further include a memory 309 arranged to store, for example, a first light output spectrum, an adjusted first light output spectrum, a brightness contrast metric of the captured first image, a threshold, computer vision and / or machine learning algorithms, optical characteristics of substrates 153, 253 and / or multiple components 155, 255a-h, 257a-c, historical data, a component type library, a lookup table, etc. The memory 309 may include one or more suitable memory devices, such as one or more random access memories (RAM), read-only memories (ROM), dynamic random access memories (DRAM), fast cycle RAM (FCRAM), static RAM (SRAM), field-programmable gate arrays (FPGAs), erasable programmable read-only memories (EPROMs), electrically erasable programmable read-only memories (EEPROMs), microcontrollers, or microprocessors.

[0056] The controller 310 may be implemented in a unit separate from the light source 120 and / or imaging unit 130—such as a wall panel, a desktop computer terminal, or even a portable terminal (such as a laptop, tablet, or smartphone). Alternatively, the controller 310 may be incorporated into the same unit as the light source 120 and / or imaging unit 130. Furthermore, the controller 310 may be implemented remotely (e.g., on a server at different geographical locations); and the controller 310 may be implemented in a single unit or as a distributed function distributed across multiple independent units (e.g., a distributed server comprising multiple server units at one or more geographical locations, or distributed control functions distributed across the light source 120 and imaging unit 130). Additionally, the controller 310 may be implemented as software stored in memory (including one or more memory devices) and arranged for execution on a processor (including one or more processing units), or the controller 310 may be implemented as dedicated hardware circuitry, or configurable or reconfigurable circuitry such as a PGA or FPGA, or any combination thereof.

[0057] Regarding the various communications involved in implementing this function—for example, enabling the controller 310 to obtain, for example, the optical characteristics of the substrate 153 and / or multiple components 155, 255a-h, 157a-c, or to control the light output of the light source 120—these can be implemented by any suitable wired and / or wireless means, such as by means of: wired networks, such as Ethernet, DMX networks, or the Internet; or wireless networks, such as local (short-range) RF networks (e.g., Wi-Fi, ZigBee, or Bluetooth networks); or any combination of these and / or other means.

[0058] Figure 4 A flowchart is schematically and exemplaryly illustrated, showing an embodiment of a method 400 for inspecting the lighting device 150 during the assembly process of the lighting device 150. Method 400 may include the step of illuminating 410 of the lighting device 150 according to a first light output spectrum to provide brightness contrast between the substrate 153 and the plurality of components 155, 255a-h, 257a-c. A controller 110 may be used to control the illumination of the light source 120. Method 400 may further include capturing 420 a first image of the illuminated lighting device 150. An imaging unit 130 (e.g., a camera) may be arranged to capture the first image. The first image may be a still image.

[0059] Method 400 may include determining a brightness contrast metric of the first image captured by 430. Controller 110 may be arranged to determine whether the brightness contrast metric of the first image captured by 430 exceeds a threshold. If the brightness contrast metric exceeds the threshold 440b, method 400 may include adjusting the first light output spectrum 450b. Controller 110 may be arranged to adjust the first light output spectrum 450b. Adjustment 450b may be performed based on the optical characteristics of substrate 153 and / or multiple components 155, 255a-h, 257a-c and / or based on scanning different spectra. In this example, the first light output spectrum is suitable for one or more of the multiple components 155, 255a-h, 257a-c. In other words, different spectra may be suitable for different components. In this case, different (suitable) spectra may be used.

[0060] Method 400 may further include capturing 460b a second image of the lighting device 150 illuminated according to the adjusted first light output spectrum. The second image may be captured by imaging unit 130. The method may then further include comparing the second image with a reference image 470b. The reference image includes an image of the lighting device without any defects(s). The comparison 470b may include using computer vision and / or machine learning algorithms. Method 400 may further include determining, based on the comparison, defects in substrate 153 and / or in multiple components 155, 255a-h, 257a-c 480b.

[0061] When the brightness contrast metric of the captured first image does not exceed the threshold of 440a, method 400 may then include comparing the first image with a reference image 450a, and then determining, based on the comparison, defects in substrates 153, 253 and / or multiple components 155, 255a-h, 257a-c.

[0062] When the computer program product is run on the processing unit of the computing device (such as the processing unit 303 of the controller 110), the method 400 can be executed by the computer program code of the computer program product.

[0063] It should be noted that the above embodiments are illustrative and not limiting of the invention, and those skilled in the art will be able to devise many alternative embodiments without departing from the scope of the appended claims.

[0064] In the claims, any reference numerals placed between parentheses should not be construed as limiting the claims. The use of the verb "comprising" and its variations does not exclude the presence of elements or steps other than those stated in the claims. The article "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer or processing unit. In an apparatus claim enumerating several means, several of these means may be embodied by the same item of hardware. The mere fact that certain measures are referenced in mutually different dependent claims does not indicate that a combination of these measures cannot be used advantageously.

[0065] The various aspects of this invention can be implemented in a computer program product, which may be a collection of computer program instructions stored on a computer-readable storage device and executable by a computer. The instructions of this invention can be any interpretable or executable code mechanism, including but not limited to scripts, interpretable programs, dynamic link libraries (DLLs), or Java classes. The instructions may be provided as a complete executable program, a partial executable program, as a modification (e.g., an update) of an existing program, or as an extension (e.g., a plugin) of an existing program. Furthermore, portions of the processing of this invention may be distributed across multiple computers or processors or even the “cloud.”

[0066] Storage media suitable for storing computer program instructions include all forms of non-volatile memory, including but not limited to EPROM, EEPROM, and flash memory devices, disks such as internal and external hard drives, removable disks, and CD-ROMs. Computer program products may be distributed on such storage media or made available for download via HTTP, FTP, email, or through a server connected to a network such as the Internet.

Claims

1. An inspection system for inspecting a lighting device during an assembly process of the lighting device; wherein the lighting device comprises a substrate and a plurality of components mounted on the substrate; wherein the inspection system comprises: - a light source arranged for illuminating the lighting device according to a first light output spectrum to provide a luminance contrast between the substrate and the plurality of components; - an imaging unit arranged for capturing a first image of the illuminated lighting device; - a controller comprising a processing unit for determining a luminance contrast measure of the captured first image; wherein the processing unit is further arranged for adjusting the first light output spectrum when the luminance contrast measure of the captured first image exceeds a threshold value, and wherein the imaging unit is further arranged for capturing a second image of the lighting device illuminated according to the adjusted first light output spectrum; and wherein the controller further comprises: - a comparison unit arranged for comparing the second image to a reference image; - a determination unit arranged for determining a defect in the substrate and / or in the plurality of components based on the comparison.

2. The inspection system according to claim 1; wherein the comparison unit is arranged for comparing the first image to the reference image when the luminance contrast measure of the captured first image does not exceed the threshold value.

3. The inspection system according to claim 1; wherein the processing unit is arranged for: - obtaining a signal indicative of an optical property of the substrate and / or the plurality of components; - adjusting the first light output spectrum based on the obtained optical property.

4. The inspection system according to claim 1; wherein the processing unit is arranged for adjusting the first light output spectrum by sequentially increasing or decreasing the first light output spectrum.

5. The inspection system according to claim 1; wherein the lighting device is not powered during the assembly process.

6. The inspection system according to claim 1; wherein the plurality of components comprises at least phosphor-coated LEDs.

7. The inspection system according to claim 1; wherein the comparison unit is arranged for image comparison using computer vision and / or machine learning algorithms.

8. The inspection system according to claim 7; wherein the threshold value is determined by computer vision and / or machine learning algorithms.

9. The inspection system according to claim 1; wherein the light source is a multi-spectral light source arranged for illuminating the lighting device according to at least the first light output spectrum and a second spectrum; wherein the processing unit is further arranged for adjusting the first light output spectrum and / or the second spectrum when a luminance contrast measure of an image of the lighting device illuminated by the multi-spectral light source exceeds a threshold value.

10. The inspection system according to claim 9; wherein the imaging unit is a multi-spectral sensing device.

11. The inspection system of claim 1; wherein the imaging unit comprises a thermal imaging device arranged for capturing a first thermal image of the illuminated lighting device illuminated according to the first light output spectrum and / or a second thermal image of the illuminated lighting device illuminated according to the adjusted first light output spectrum.

12. The inspection system of claim 11; wherein the comparison unit is further arranged for comparing the first thermal image or the second thermal image to a reference thermal image; and the determination unit is further arranged for determining a defect in the substrate and / or in the plurality of components based on the comparison.

13. A controller for use with the inspection system of claim 1 to inspect a luminaire during assembly of the luminaire. wherein the controller comprises: - an input interface and an output interface; - a processing unit arranged for determining a luminance contrast measure of a first image of the lighting device illuminated according to a first light output spectrum; and wherein the processing unit is further arranged for adjusting the first light output spectrum when the luminance contrast measure of the captured first image exceeds a threshold value; - a comparison unit arranged for comparing the first image or a second image of the lighting device illuminated according to the adjusted first light output spectrum to a reference image; - a determination unit arranged for determining a defect in the substrate and / or in the plurality of components of the lighting device based on the comparison.

14. A method for inspecting a lighting device during an assembly process of the lighting device with the inspection system of claim 1; wherein the lighting device comprises a substrate and a plurality of components mounted on the substrate; wherein the method comprises the steps of: - illuminating the lighting device according to a first light output spectrum to provide a luminance contrast between the substrate and the plurality of components; - capturing a first image of the illuminated lighting device; - determining a luminance contrast measure of the captured first image; and wherein the first light output spectrum is adjusted when the luminance contrast measure of the captured first image exceeds a threshold value; and a second image of the lighting device illuminated according to the adjusted first light output spectrum is captured; wherein the method further comprises: - comparing the second image to a reference image, - determining a defect in the substrate and / or in the plurality of components based on the comparison.

15. A computer program product comprising instructions which, when the program is executed by a computer, cause the computer to carry out the steps of the method of claim 14.

Images