System and method for matching color and appearance of target coating

Abstract

The present invention relates to systems and methods for matching the color and appearance of a target coating. The systems and methods include receiving target image data associated with a target coating. A feature extraction analysis process is applied to the target image data to determine target image features. The feature extraction analysis process includes dividing the target image into sub-images containing a plurality of target pixels. A machine learning model uses the target pixel features to identify one or more types of flakes present in the target coating.

Description

[0001] Cross-references to related applications

[0002] This application claims priority to U.S. Provisional Patent Application No. 62 / 955,732, filed December 31, 2019, the entire contents of which are incorporated herein by reference. Technical Field

[0003] The technical field relates to coating technology, and more specifically, to systems and methods for matching the color and appearance of a target coating. Background Technology

[0004] The visualization and selection of coatings with desired colors and appearances play a crucial role in many applications. For example, paint suppliers must offer thousands of paints to cover the range of coatings used by global OEMs for all current and latest vehicle models. Offering such a large number of different paints as factory-packaged products increases the complexity of paint manufacturing and inventory costs. Therefore, paint suppliers provide mixing systems for paint formulations that typically include 50 to 100 components (e.g., single pigment colors, binders, solvents, additives) and components that match the range of coatings available for the vehicle. The mixing system can be located at a repair shop (i.e., a body shop) or paint dealer and allows users to obtain a coating with the desired color and appearance by dispensing the components in amounts corresponding to the paint formulation. Paint formulations are typically maintained in a database and distributed to customers via computer software through download or direct connection to an internet database. Each paint formulation typically involves one or more alternative paint formulations to account for coating variations due to changes in vehicle production.

[0005] Identifying the paint formulation most similar to the target coating is complicated by this variation. For example, a particular coating might appear on three vehicle models manufactured in two assembly plants with different application equipment using paints from two different OEM paint suppliers and exceeding five years of vehicle life. These sources of variation result in significant coating variations within the group of vehicles with that particular coating. Alternative paint formulations provided by paint suppliers are matched to a subset of the color group, ensuring a close match for any vehicle requiring repair. Each alternative paint formulation can be represented by a color chart in a bar chart, allowing users to select the best matching formulation through visual comparison with the vehicle.

[0006] Identifying the paint formulation most similar to the target coating for repair is typically accomplished by using a spectrophotometer or a fandeck. The spectrophotometer measures one or more color and appearance attributes of the target coating to be repaired. This color and appearance data is then compared to corresponding data from possible candidate formulations contained in a database. The candidate formulation whose color and appearance attributes best match the color and appearance attributes of the target coating to be repaired is then selected as the paint formulation most similar to the target coating. However, spectrophotometers are expensive and not readily available in the economic market.

[0007] Alternatively, the fandeck includes a plurality of sample coatings on pages or sheets within the fandeck. The sample coatings of the fandeck are then visually compared to the target coating being repaired. The formulation associated with the sample coating that best matches the color and appearance attributes of the target coating to be repaired is then selected as the paint formulation most similar to the target coating. However, fandecks are cumbersome to use and difficult to maintain due to the large number of sample coatings needed to account for all coatings on vehicles on the road today.

[0008] Accordingly, it is desirable to provide a system and method for matching the color and appearance of a target coating. In addition, other desirable features and characteristics will become apparent from the subsequent and specific embodiments, as well as the appended claims, taken in conjunction with the accompanying drawings and this background. SUMMARY

[0009] Various non-limiting embodiments of a system for matching the color and appearance of a target coating and various non-limiting embodiments of a method for matching the color and appearance of a target coating are disclosed herein. In one non-limiting embodiment, the system includes, but is not limited to, a storage device for storing instructions and one or more data processors. The one or more data processors are configured to receive a target image of a target coating. The target image includes target image data. A feature extraction analysis process is applied to the target image data to determine target image features. The feature extraction analysis process includes dividing the target image into sub-images containing a plurality of target pixels. The sub-images include or do not include flakes. Target pixel features of the sub-images are determined. A machine learning model is applied to identify one or more types of flakes present in the target coating using the determined target pixel features.

[0010] A system and method can also include receiving target image data associated with a target coating. A feature extraction analysis process is applied to the target image data to determine target image features. The feature extraction analysis process includes dividing the target image into sub-images containing a plurality of target pixels. A machine learning model identifies one or more types of flakes present in the target coating using the target pixel features. BRIEF DESCRIPTION OF DRAWINGS

[0011] Other advantages of the disclosed subject matter will be readily appreciated, as the same becomes better understood by reference to the following detailed description when considered in connection with the accompanying drawings wherein:

[0012] Figure 1 is a perspective view showing a non-limiting embodiment of a system for matching a color and appearance of a target coating;

[0013] Figure 2 is a block diagram showing a non-limiting embodiment of a system of Figure 1

[0014] Figure 3A is an image showing a non-limiting embodiment of a target coating of Figure 1

[0015] Figure 3B is a graphical representation of RGB values of a non-limiting embodiment of a target coating of Figure 3A

[0016] Figure 4A is an image showing a non-limiting embodiment of a first sample image of a system of Figure 1

[0017] Figure 4B is a graphical representation of RGB values of a non-limiting embodiment of a first sample image of Figure 4A

[0018] Figure 5A is an image showing a non-limiting embodiment of a second sample image of a system of Figure 1

[0019] Figure 5B is a graphical representation of RGB values of a non-limiting embodiment of a second sample image of Figure 5A

[0020] Figure 6 is a perspective view showing a non-limiting embodiment of an electronic imaging device of a system of Figure 1

[0021] Figure 7 is another perspective view showing a non-limiting embodiment of an electronic imaging device of a system of Figure 1

[0022] Figure 8 is a flowchart showing a non-limiting embodiment of a system of Figure 1

[0023] Figure 9 is a flowchart showing a non-limiting embodiment of a method of Figure 8 ​​​​​​​​​​​

[0024] Figure 10 is a flowchart illustrating another non-limiting embodiment of a method of Figure 8

[0025] Figure 11 is a schematic diagram illustrating the formation of a cross-sectional portion of a substrate and coating layer;

[0026] Figure 12 and Figure 13 is a schematic diagram illustrating the capturing of an image of a coating layer;

[0027] Figure 14 and Figure 15 is a flowchart illustrating an embodiment of a system and method;

[0028] Figure 16 is a hypothetical diagram illustrating one possible portion of a technique for determining target and / or sample features; and

[0029] Figures 17-19 is a flowchart illustrating an embodiment of a method.

[0030] Figure 20 is a schematic diagram illustrating the application of a repair coating to a substrate.

[0031] Figures 21-23 depicts a color matching analysis system for analyzing one or more features of a target coating.

[0032] Figure 24 depicts a machine learning model applied to a distribution of target coating features and statistics.

[0033] Figure 25 depicts a color analysis expert system that applies color analysis expert rules to output patch prediction probabilities.

[0034] Figure 26 depicts a system for assisting a colorist in understanding and correcting for differences in spatial color.

[0035] Figure 27 depicts a process flow for determining when a colorist achieves an acceptable match.

[0036] Figure 28 depicts a system for analyzing a target coating for texture matching.

[0037] Figure 29 depicts a server environment in which a user can interact with a color and appearance matching analysis system. DETAILED DESCRIPTION

[0038] ​The following detailed description includes examples and does not intend to limit, or be

[0039] Those of ordinary skill in the art will more readily appreciate the features and advantages of the present disclosure, together with others, upon reading the following detailed description that presents embodiments that are disclosed. It should be appreciated that certain features and subcombinations are of utility and can be employed without reference to other features and subcombinations. This is contemplated by and is within the scope of the claims. It will be readily understood that the components of the present disclosure, as generally described and illustrated in the Figures herein, could be arranged and designed in a wide variety of different configurations. Thus, the following more detailed description, as represented in the Figures, is not intended to limit the scope, application or uses of any embodiments of the disclosure. Instead, they are presented as example configurations provided to

[0040] Unless expressly identified otherwise, the use of numerical values in the various ranges specified in the present disclosure are specified as approximations. In this manner, slight variations above and below the stated ranges can be used to achieve substantially the same result as the stated range without rendering such modifications to the range outside the scope of the disclosure. Moreover, these ranges have been disclosed to provide a range of alternatives for one of ordinary skill in the art to implement. Unless expressly indicated otherwise, as with the minimum and maximum values specified in a range, the use of numerical values in the various ranges specified in the present disclosure are specified as approximations. In this manner, slight variations above and below the stated ranges can be used to achieve substantially the same result as the stated range without rendering such modifications to the range outside the scope of the disclosure. Moreover, these ranges have been disclosed to provide a range of alternatives for one of ordinary skill in the art to implement.

[0041] In this document, processes and techniques can be described in terms of functional and / or logical blocks components and with reference to symbolic representations of operations, processing tasks, and functions that can be performed by various computing components or devices. It will be appreciated that various block components can be implemented by any number of hardware, software, and / or firmware components configured to perform the specified functions. For example, an embodiment of a system or a component can employ various integrated circuit components, e.g., memory elements, digital signal processing elements, logic elements, look-up tables, or the like, which can be configured to perform one or more functions and operations.

[0042] The following description can refer to elements or nodes or features being "coupled" together. As used herein, unless expressly stated to the contrary, "coupled" means that one element / node / feature is directly or indirectly joined to another element / node / feature, and can not necessarily mean that the two elements / nodes / features are in direct physical or electrical contact. As such, although the drawings can depict one example of an arrangement of elements, additional intervening elements, devices, features, or components can be present in an embodiment of the described subject matter. Additionally, certain terminology can also be used in the following description for the purpose of reference only, and, therefore, is not intended to be limiting.

[0043] In this document, processes and techniques can be described in terms of functional and / or logical block components and in terms of symbolic representations of operations, processing tasks, and functions. These operations, tasks, and functions can sometimes be referred to as computer-executed, computerized, software-implemented, or computer-implemented. In practice, one or more processor devices can perform the described operations, tasks, and functions through manipulations of electrical signals representing data bits stored in system memory. The storage locations in system memory are physical locations having particular electrical, magnetic, optical, or organic properties structured to store data bits. It should be appreciated that the various block components shown in the figures can be realized by any number of hardware, software, and / or firmware components configured to perform the specified functions. For example, an embodiment of a system or component can employ various integrated circuit components, e.g., memory elements, digital signal processing elements, logic elements, look-up tables, or the like, which can carry out a variety of functions under the control of one or more microprocessors or other control devices.

[0044] For the sake of brevity, conventional techniques related to graphic and image processing, touch screen displays, and other functional aspects of the systems (and the individual operating components of the systems) can not be described in detail herein. Furthermore, the connecting lines shown in the various figures included herein are intended to represent example functional relationships and / or physical couplings between the various elements. It should be noted that many alternatives or additional functional relationships or physical connections can be present in an embodiment of the subject matter.

[0045] As used herein, the term "module" refers to any hardware, software, firmware, electronic control component, processing logic, and / or processor device, alone or in any combination, including but not limited to: application-specific integrated circuits (ASICs), electronic circuits, shared, dedicated, or group shared, processors and memory that execute one or more software or firmware programs, a combination of logic circuits and / or other suitable components that provide the described functionality.

[0046] As used herein, the term "pigment" or "plurality of pigments" refers to one or more colorants that produce one or more colors. Pigments can be from natural or synthetic sources and made from organic or inorganic ingredients. Pigments can also include metal particles or flakes having specific or mixed shapes and sizes. Pigments are generally insoluble in the coating composition.

[0047] The term "effect pigment" or "plurality of effect pigments" refers to pigments that produce special effects in a coating layer. Examples of effect pigments include, but are not limited to, light scattering pigments, light interference pigments, and light reflecting pigments. Metal flakes, such as aluminum flakes, and pearlescent pigments, such as mica-based pigments, are examples of effect pigments.

[0048] The term "appearance" can include (1) aspects of the visual experience of observing or recognizing a coating; and (2) the perception of combining the spectral and geometric aspects of a coating with its illumination and viewing environment. Generally, appearance includes the texture, granularity, sparkle, or other visual effects of a coating, particularly when viewed from different angles of observation and / or under different illumination conditions. Appearance characteristics or appearance data can include, but are not limited to, descriptive or measured data regarding texture, metallic effect, pearlescent effect, gloss, image clarity, flake appearance and size (e.g., texture, granularity, sparkle, luminescence and scintillation, and depth perception enhancement in a coating imparted by flakes, particularly by metallic flakes such as aluminum flakes). Appearance characteristics can be obtained by visual inspection or by using an appearance measurement device.

[0049] The term "color data" or "color characteristics" of a coating can include measured color data including spectral reflectance values, C, U, Z values, L*, a*, b* values, L*, a*, b* values, L, C, h values, or combinations thereof. Color data can also include a color code, color name or description of a vehicle, or combinations thereof. Color data can even include visual aspects of a color, chroma, hue, lightness or darkness of a coating. Color data can be obtained by visual inspection or by using a color measurement device such as a colorimeter, spectrophotometer, or gonio-spectrophotometer. In particular, a spectrophotometer obtains color data by determining the wavelengths of light reflected by a coating. Color data can also include descriptive data such as a name of a color, a color code of a vehicle; a binary, textural or encrypted data file containing a description of one or more colors; a measured data file such as a measured data file generated by a color measurement device; or an export / import data file generated by a computing device or color measurement device. Color data can also be generated by an appearance measurement device or a color-appearance dual measurement device.

[0050] The term "coating" or "coating composition" can include any coating composition known to those skilled in the art and can include a two-component coating composition, also known as a "2K coating composition"; a one-component or IK coating composition; a coating composition having a cross-linkable component and a cross-linking component; a radiation-curable coating composition such as a UV-curable coating composition or an E-beam curable coating composition; a single-cure coating composition; a dual-cure coating composition; a lacquer coating composition; a waterborne coating composition or an aqueous coating composition; a solvent-borne coating composition; or any other coating composition known to those skilled in the art. A coating composition can be formulated as a primer, basecoat, or pigmented coating composition by incorporating a desired pigment or effect pigment. A coating composition can also be formulated as a clear coating composition.

[0051] The terms "vehicle," "automobile," "motor vehicle," or "motorized vehicle" can include motorized vehicles such as sedans, buses, trucks, semi-trucks, vans, SUVs (Sport Utility Vehicles); tractors; motorcycles; trailers; ATVs (All Terrain Vehicles); heavy machinery such as bulldozers, mobile cranes, and excavators; airplanes; boats; ships; and other modes of transportation.

[0052] The terms "recipe," "matching recipe," or "matching regimen" for a coating composition refer to a collection of information or instructions based on which the coating composition can be prepared. In one example, the matching recipe includes a list of names and amounts of pigments, effect pigments, and other components of the coating composition. In another example, the matching recipe includes instructions on how to mix the various components of the coating composition.

[0053] Reference is made herein to Figure 1 A processor-implemented system 10 is provided for matching the color and appearance of a target coating 12. The target coating 12 can be on a substrate 14. The substrate 14 can be a vehicle or a portion of a vehicle. The substrate 14 can also be any coated article that includes the target coating 12. The target coating 12 can include a pigmented coating, a clear coating, or a combination of a pigmented coating and a clear coating. The pigmented coating can be formed from a pigmented coating composition. The clear coating can be formed from a clear coating composition. The target coating 12 can be formed from one or more solvent-borne coating compositions, one or more water-borne coating compositions, one or more two-component coating compositions, or one or more one-component coating compositions. The target coating 12 can also be formed from one or more coating compositions each having a cross-linkable component and a cross-linking component, one or more radiation-curable coating compositions, or one or more lacquer coating compositions.

[0054] Reference is made to Figure 2 and with continued reference to Figure 1 The system 10 includes an electronic imaging device 16 configured to generate target image data 18 of the target coating 12. The electronic imaging device 16 can be a device capable of capturing images at a wide range of electromagnetic wavelengths including visible wavelengths or non-visible wavelengths. The electronic imaging device 16 can also be defined as a mobile device. Examples of mobile devices include, but are not limited to, a mobile phone (e.g., a smart phone), a mobile computer (e.g., a tablet computer or a laptop computer), a wearable device (e.g., a smart watch or earpiece), or any other type of device known in the art configured to receive the target image data 18. In one implementation, the mobile device is a smart phone or a tablet computer.

[0055] In implementations, the electronic imaging device 16 includes a video camera 20 (see Figure 7). The camera 20 can be configured to acquire target image data 18. The camera 20 can be configured to capture images having visible wavelengths. The target image data 18 can be derived from a target image 58 of the target coating 12, such as a still image or a video. In certain embodiments, the target image data 18 is derived from a still image. In Figure 1 In the illustrated embodiment, the electronic imaging device 16 is shown disposed proximate to and spaced apart from the target coating 12. However, it should be appreciated that the electronic imaging device 16 of the embodiments is portable such that the electronic imaging device 16 can be moved to another coating (not shown). In other embodiments (not shown), the electronic imaging device 16 can be fixed at one location. In still other embodiments (not shown), the electronic imaging device 16 can be attached to a robotic arm for automated movement. In further embodiments (not shown), the electronic imaging device 16 can be configured to measure properties of multiple surfaces simultaneously.

[0056] The system 10 also includes a storage device 22 for storing instructions for performing matching of the color and appearance of the target coating 12. The storage device 22 can store instructions that are executable by the one or more data processors 24. The instructions stored in the storage device 22 can include one or more separate programs, each of which comprises an ordered listing of executable instructions for implementing logical functions. The one or more data processors 24 are configured to, when the system 10 is in operation, execute instructions stored in the storage device 22 to transfer data to and from the storage device 22, and, in general, to

[0057] As introduced above, the system 10 also includes one or more data processors 24 configured to execute instructions. The one or more data processors 24 are configured to be communicatively coupled with the electronic imaging device 16. The one or more data processors 24 can be any custom or commercially available processor, central processing unit (CPU), auxiliary processor, semiconductor-based microprocessor (in the form of a microchip or chip set), or general any device for executing instructions associated with the electronic imaging device 16. The one or more data processors 24 can be communicatively coupled with any component of the system 10 by a wired connection, a wireless connection, and / or a device or combination thereof. Examples of suitable wired connections include, but are not limited to, a hardware coupling, a separator, a connector, a cable, or a wire. Examples of suitable wireless connections and devices include, but are not limited to, a Wi-Fi device, a Bluetooth device, a wide area network (WAN) wireless device, a Wi-Max device, a local area network (LAN) device, a 3G broadband device, an infrared communication device, an optical data transfer device, a radio transmitter and optional receiver, a wireless telephone, a wireless telephone adapter card, or any other device that can transmit signals at a wide range of electromagnetic wavelengths including radio frequencies, microwave frequencies, visible wavelengths, or non-visible wavelengths.

[0058] With reference to Figure 3A and Figure 3B The one or more data processors 24 are configured to execute instructions to receive target image data 18 of the target coating 12 by the one or more data processors 24. As described above, the target image data 18 is generated by the electronic imaging device 16. The target image data 18 can define RGB values, L*a*b* values, or a combination thereof representing the target coating 12. In certain embodiments, the target image data 18 defines RGB values representing the target coating 12. The one or more data processors 24 can also be configured to execute instructions to convert the RGB values of the target image data 18 to L*a*b* values representing the target coating 12.

[0059] The target image data 18 includes target image features 26. The target image features 26 can include color and appearance characteristics of the target coating 12, a representation of the target image data 18, or a combination thereof. In certain embodiments, the target image features 26 can include a representation based on image entropy.

[0060] The one or more data processors 24 are configured to execute instructions to retrieve, by the one or more data processors 24, one or more feature extraction analysis processes 28’ that extract target image features 26 from the target image data 18. In implementations, the one or more feature extraction analysis processes 28’ are configured to identify an image entropy-based representation to extract the target image features 26 from the target image data 18. To this end, the one or more data processors 24 can be configured to execute instructions to identify an image entropy-based representation to extract the target image features 26 from the target image data 18.

[0061] Identifying the image entropy-based representation can include determining a color image entropy curve of the target image data 18. The target image data 18 can be represented in a three-dimensional L*a*b* space with a color entropy curve based on a Shannon entropy of each a*b* plane, each L*a*b* plane, or a combination thereof. Determining the color entropy curve can include partitioning the three-dimensional L*a*b* space of the target image data 18 into a plurality of cubic subspaces; tabulating cubic spaces with similar characteristics to arrive at a total cubic space count for each characteristic; generating an empty image entropy array for each dimension of the three-dimensional L*a*b* space; and populating the empty image entropy array with the total cubic space count corresponding to each dimension.

[0062] Identifying the image entropy-based representation can also include determining a color difference image entropy curve of the target image data 18. The target image data 18 can be represented in a three-dimensional L*a*b* space with a color difference entropy curve based on a three-dimensional L*a*b* space analysis of an alternative three-dimensional L*a*b* space. Determining the color difference entropy curve can include calculating a dL* image entropy, a dC* image entropy, and a dh* image entropy between the three-dimensional L*a*b* space and the alternative three-dimensional L*a*b* space.

[0063] Identifying the image entropy-based representation can also include determining a black and white intensity image entropy of an L* plane from the three-dimensional L*a*b* space of the target image data 18. Identifying the image entropy-based representation can also include determining an average L*a*b* value of the target image data 18. Identifying the image entropy-based representation can also include determining an L*a*b* value of a center of a most dense cubic subspace.

[0064] The one or more data processors 24 are also configured to execute the above-described instructions to apply the target image data 18 to the one or more feature extraction analysis processes 28’. The one or more data processors 24 are also configured to execute the above-described instructions to extract the target image features 26 from the target image data 18 with the one or more feature extraction analysis processes 28’.

[0065] In implementations, the system 10 is configured to extract the target image feature 26 from the target image data 18 by identifying an image entropy-based representation of the target image feature 26. Identifying the image entropy-based representation can include determining a color image entropy curve of the target image data 18, determining a color image entropy curve of the target image data 18, determining a black and white intensity image entropy of an L* plane from a three-dimensional L*a*b* space of the target image data 18, determining an average L*a*b* value of the target image data 18, determining an L*a*b* value of a center of a densest cubical subspace, or a combination thereof.

[0066] Referring to Figure 4A and Figure 5A and continuing to refer to Figure 2 In implementations, the system 10 further includes a sample database 30. The sample database 30 can be associated with the electronic imaging device 16, for example, in a server-based environment or a cloud computing environment, or separate from the electronic imaging device 16. It should be appreciated that the one or more data processors 24 are configured to be communicatively coupled with the sample database 30. The sample database 30 can include a plurality of sample images 32, such as a first sample image 34 shown in Figure 4A and a second sample image 36 shown in Figure 5A In implementations, each sample image of the plurality of sample images 32 is an image of a panel including a sample coating. Various sample coatings defining a set of paint formulations can be imaged to generate the plurality of sample images 32. The sample images 32 can be imaged with one or more different electronic imaging devices 16 to account for variations in imaging capabilities and performance of each electronic imaging device 16. The plurality of sample images 32 can be in any format, such as RAW, JPEG, TIFF, BMP, GIF, PNG, and the like.

[0067] The one or more data processors 24 can be configured to execute instructions to receive sample image data 38 of the sample images 32 by the one or more data processors 24. The sample image data 38 can be generated by the electronic imaging device 16. The sample image data 38 can define RGB values, L*a*b* values, or a combination thereof, representing the sample images 32. In certain implementations, the sample image data 38 defines RGB values representing the sample images 32, such as the RGB values of the first sample image 34 shown in Figure 4B and the RGB values of the second sample image 36 shown in Figure 5B The one or more data processors 24 can be further configured to execute instructions to convert the RGB values of the sample image data 38 to L*a*b* values representing the sample images 32. The system 10 can be configured to normalize the sample image data 38 of the plurality of sample images 32 for various electronic imaging devices 16, thereby improving performance of the system 10.

[0068] The sample image data 38 can include sample image features 40. The sample image features 40 can include color and appearance characteristics of the sample image 32, a representation of the sample image data 38, or a combination thereof. In certain embodiments, the sample image features 40 can include an image entropy-based representation.

[0069] The one or more data processors 24 are configured to execute instructions to retrieve, by the one or more data processors 24, one or more feature extraction analysis processes 28” that extract the sample image features 40 from the sample image data 38. In embodiments, the one or more feature extraction analysis processes 28” are configured to identify an image entropy-based representation to extract the sample image features 40 from the sample image data 38. To this end, the one or more data processors 24 can be configured to execute instructions to identify an image entropy-based representation to extract the sample image features 40 from the sample image data 38. It will be appreciated that the one or more feature extraction analysis processes 28” used to extract the sample image features 40 can be the same or different than the one or more feature extraction analysis processes 28’ used to extract the target image features 26.

[0070] In embodiments, the system 10 is configured to extract the sample image features 40 from the sample image data 38 by identifying an image entropy-based representation of the sample image features 40. Identifying an image entropy-based representation can include determining a color image entropy curve of the sample image data 38; determining a color image entropy curve of the sample image data 38; determining a black and white intensity image entropy of an L* plane from a three-dimensional L*a*b* space of the sample image data 38; determining an average L*a*b* value of the sample image data 38; determining an L*a*b* value of a center of a densest cubical subspace; or a combination thereof.

[0071] The one or more data processors 24 are configured to execute instructions to retrieve, by the one or more data processors, a machine learning model 42 that utilizes the target image features 26 to identify a matching sample image 44 from the plurality of sample images 32 that is calculated. The machine learning model 42 can employ supervised training, unsupervised training, or a combination thereof. In embodiments, the machine learning model 42 employs supervised training. Examples of suitable machine learning models include, but are not limited to, linear regression, decision trees, k-means clustering, principal component analysis (PCA), random decision forests, neural networks, or any other type of machine learning algorithm known in the art. In embodiments, the machine learning model is based on a random decision forest algorithm.

[0072] The machine learning model 42 includes pre-specified matching criteria 46 representing the plurality of sample images 32 for identifying the computed matching sample image 44 from the plurality of sample images 32. In embodiments, the pre-specified matching criteria 46 are arranged in one or more decision trees. The one or more data processors 24 are configured to apply the target image features 26 to the machine learning model 42. In embodiments, the pre-specified matching criteria 46 are included in one or more decision trees, where the decision trees include root nodes, intermediate nodes throughout different levels, and end nodes. The target image features 26 can be processed through the nodes to one or more end nodes, where each end node represents one of the plurality of sample images 32.

[0073] The one or more data processors 24 are further configured to identify the computed matching sample image 44 based on substantially satisfying the one or more pre-specified matching criteria 46. In embodiments, the phrase“substantially satisfying” means identifying the computed matching sample image 44 from the plurality of sample images 32 with the greatest probability of matching the target coating 12. In embodiments, the machine learning model 42 is based on a random decision forest algorithm including a plurality of decision trees, where the result of the processing of the target image features 26 through each decision tree is used to determine the probability of each sample image 32 matching the target coating 12. The sample image 32 with the greatest probability of matching the target coating 12 can be defined as the computed matching sample image 44.

[0074] In embodiments, the one or more data processors 24 are configured to execute instructions to generate the pre-specified matching criteria 46 of the machine learning model 42 based on the sample image features 40. In certain embodiments, the pre-specified matching criteria 46 are generated based on the sample image features 40 extracted from the plurality of sample images 32. The one or more data processors 24 can be configured to execute instructions to train the machine learning model 42 based on the plurality of sample images 32 by generating the pre-specified matching criteria 46 based on the sample image features 40. The machine learning model 42 can be trained at regular intervals (e.g., monthly) based on the plurality of sample images 32 included in the sample database 30. As described above, the sample image data 38 defining the RGB values representing the sample images 32 can be converted to L*a*b* values with the sample image features 40 extracted from the sample image data 38 including L*a*b* values by identifying the representation based on image entropy.

[0075] The computed matching sample image 44 is utilized to match the color and appearance of the target coating 12. The computed matching sample image 44 can correspond to a paint formulation that can match the color and appearance of the target coating 12. The system 10 can include one or more alternative matching sample images 48 that are related to the computed matching sample image 44. The one or more alternative matching sample images 48 can be related to the computed matching sample image 44 based on paint formulations, observed similarities, computed similarities, or combinations thereof. In certain embodiments, the one or more alternative matching sample images 48 are related to the computed matching sample image 44 based on paint formulations. In embodiments, the computed matching sample image 44 corresponds to a primary paint formulation and the one or more alternative matching sample images 48 correspond to alternative paint formulations related to the primary paint formulation. The system 10 can include a visual matching sample image 50 that can be selected by a user from the computed matching sample image 44 and the one or more alternative matching sample images 48 based on similarities observed by the user to the target coating 12.

[0076] Referring to Figure 6 and Figure 7 In embodiments, the electronic imaging device 16 further includes a display 52 configured to display the computed matching sample image 44. In certain embodiments, the display 52 is further configured to display a target image 58 of the target coating 12 adjacent to the computed matching sample image 44. In embodiments, the display 52 is further configured to display the one or more alternative matching sample images 48 related to the computed matching sample image 44. In embodiments of the electronic imaging device 16 that include the video camera 20, the display 52 can be located across from the video camera 20.

[0077] In embodiments, the system 10 further includes a user input module 54 configured to select the visual matching sample image 50 by a user from the computed matching sample image 44 and the one or more alternative matching sample images 48 based on similarities observed by the user to the target coating 12. In embodiments of the electronic imaging device 16 that include the display 52, the user can select the visual matching sample image 50 through touch input on the display 52.

[0078] In embodiments, the system 10 further includes a light source 56 configured to illuminate the target coating 12. In embodiments of the electronic imaging device 16 that include the video camera 20, the electronic imaging device 16 can include the light source 56 and the light source 56 can be positioned adjacent to the video camera 20.

[0079] In embodiments, the system 10 further includes a dark box (not shown) for isolating the target coating 12 to be imaged from external light, shadows, and reflections. The dark box can be configured to house the electronic imaging device 16 and allow the target coating 12 to be exposed to the camera 20 and the light source 56. The dark box can include a light diffuser (not shown) configured to cooperate with the light source 56 to substantially diffuse light generated from the light source 56.

[0080] Reference is also made herein to Figure 8 Reference is also made herein to Figures 1-7 A method 1100 for matching a color and appearance of a target coating 12 is provided. The method 1100 includes the step 1102 of receiving, by one or more data processors, target image data 18 of the target coating 12. The target image data 18 is generated by the electronic imaging device 16 and includes the target image features 26. The method 1100 further includes the step 1104 of retrieving, by the one or more processors, one or more feature extraction analysis processes 28’ that extract the target image features 26 from the target image data 18. The method 1100 further includes the step 1106 of applying the target image features 26 to the one or more feature extraction analysis processes 28’. The method 1100 further includes the step 1108 of extracting the target image features 26 from the target image data 18 using the one or more feature extraction analysis processes 28’.

[0081] The method 1100 further includes the step 1110 of retrieving, by the one or more data processors, a machine learning model 42 that identifies a computed matching sample image 44 from the plurality of sample images 32 using the target image features 26. The machine learning model 42 includes pre-specified matching criteria 46 representing the plurality of sample images 32 for identifying the computed matching sample image 44 from the plurality of sample images 32. The method 1100 further includes the step 1112 of applying the target image features 26 to the machine learning model 42. The method 1100 further includes the step 1114 of identifying the computed matching sample image 44 based on substantially satisfying one or more of the pre-specified matching criteria 46.

[0082] In embodiments, the method 1100 further includes the step 1116 of displaying, on the display 52, the computed matching sample image 44, one or more alternative matching sample images 48 related to the computed matching sample image 44, and a target image 58 of the target coating 12 adjacent to the computed matching sample image 44 and the one or more alternative matching sample images 48. In embodiments, the method 1100 further includes the step 1118 of selecting, by a user, a visually matching sample image 50 from the computed matching sample image 44 and the one or more alternative matching sample images 48 based on an observed similarity to the target image data 18.

[0083] Referring to Figure 9 and with continued reference to Figures 1-8 In embodiments, the method 1100 further includes a step 1120 of generating the machine learning model 42 based on the plurality of sample images 32. The step 1120 of generating the machine learning model 42 can include a step 1122 of retrieving the plurality of sample images 32 from the sample database 30. The step 1120 of generating the machine learning model 42 can further include a step 1124 of extracting sample image features 40 from the plurality of sample images 32 based on the one or more feature extraction analysis processes 28’. The step 1120 of generating the machine learning model 42 can further include a step 1126 of generating the pre-designated matching criteria 46 based on the sample image features 40.

[0084] Referring to Figure 10 and with continued reference to Figures 1-9 In embodiments, the method 1100 further includes a step 1128 of forming a paint composition corresponding to the calculated matching sample image 44. The method 1100 can further include a step 1130 of applying the paint composition to the substrate 14.

[0085] The method 1100 and system 10 disclosed herein can be used for any coated article or substrate 14 including a target coating 12. Some examples of such coated articles can include, but are not limited to, household appliances such as refrigerators, washing machines, dishwashers, microwaves, cooking and baking ovens; electronic devices such as televisions, computers, electronic game consoles, audio and video equipment; recreational equipment such as bicycles, ski equipment, all-terrain vehicles; and home or office furniture such as desks, file cabinets; watercraft or marine vessels such as boats, yachts, or personal watercraft (PWC); aircraft; buildings; structures such as bridges; industrial equipment such as cranes, heavy trucks, or bulldozers; or decorative articles.

[0086] Color matching of coatings based on effect pigments is particularly challenging. Effect coatings include metallic coatings and pearlescent coatings, but can also include other effects such as phosphorescence, fluorescence, etc. As Figure 11 shown and with continued reference to Figures 1-10 Metallic coatings and pearlescent coatings include effect additives 74. Coatings including a target coating 12 and / or a sample coating 60 as shown can include several layers over a substrate 14. The target coating 12 described above can include a basecoat 70 over the substrate 14, an effect additive 74 over the basecoat 70, and a clearcoat 72 over the effect additive 74. The sample coating 60 described above can include a basecoat 70 over the substrate 14, an effect additive 74 over the basecoat 70, and a clearcoat 72 over the effect additive 74. Figure 11The illustrated sample coating 60 is identical to the layers of the target coating 12 described above. As used herein, the term "over" means that an intermediate layer can be "over" the upper component (in this example, the sample coating 60) and the lower component (in this example, the substrate 14), or means that the upper component physically contacts the lower component. Further, the term "over" means that a vertical line through the upper component also passes through the lower component, such that at least a portion of the upper component is directly above at least a portion of the lower component. It will be appreciated that the relative "over" and "under" positions can change if the substrate 14 is moved. Spatially relative terms, such as "top," "bottom," "over," and "under," are used to describe an orientation of one component relative to another component as the orientation is depicted in the figures. It will be understood that the spatially relative terms are intended to encompass different orientations of the components in addition to the orientation depicted in the figures. The components can be oriented in other orientations (e.g., rotated 90 degrees or at other Figure 11 orientations) and the spatially relative terms will be interpreted accordingly. Figure 11 Figure 11

[0087] Figure 11 A primer 62 is shown overlying the substrate 14 and a base coat 64 is shown overlying the primer 62. The primer 62 and the substrate 14 are not considered to be part of the sample coating 60 in this description. An optional effect coat 66 overlies the base coat 64, and a clear coat 68 overlies the optional effect coat 66. A sample coating formulation 70 includes a plurality of components 72, where the components 72 for the base coat 64 can not be the same as the components 72 for the optional effect coat 66 and / or the clear coat 68. One or both of the base coat 64 and the effect coat 66 include an effect additive 74 as one of the components 72. The effect additive 74 is utilized to produce a special effect for the sample coating 60, such as to produce a metallic effect or a pearlescent effect. The sample coating formulation 70 includes a base coating formulation 65 for the base coat 64, an optional effect coating formulation 67 for the optional effect coat 66, and a clear coating formulation 69 for the clear coat 68. In some embodiments, the sample coating formulation 70 can also include formulations for other optional layers.

[0088] ​​A metallic effect is produced in the case where the sample coating 60 (or any other coating) includes visible flakes. The flakes act as an effect additive 74. In embodiments, the metallic particles in the paint absorb and reflect more colors of incident light than the base paint color, giving the coating a changing appearance over a given area. Some portions of the coating will appear as the color of the base coat, while other portions will reflect light and appear as a sparkle or shimmer. Metallic colors are colors that appear as polished metallic colors. The visual typically associated with metals is a metallic luster that is different from a simple solid color. Metallic colors include a luminous effect due to the brightness of the material, and this brightness changes with the surface angle relative to the light source. One technique for producing metallic effect colors is to add aluminum flakes (which are an example of an effect additive 74) to the pigmented coating. The aluminum flakes produce a sparkle that varies in size, brightness, and sometimes color depending on the treatment of the flakes. Larger flakes produce a coarser sparkle, while smaller flakes produce a more subtle sparkle. Different types of flakes can be used, such as flat and relatively circular "dime" flakes, like a dime coin. Other types of flakes can have jagged edges, like cornflake. In some embodiments, the flakes can also be colored, so the flakes produce a colored sparkle.

[0089] Adding aluminum flakes to the base coat 64 produces a metallic effect, but if the same type and amount of flakes are added to a translucent effect coating 66 that is overlaid on the base coat 64, the coating has a "deeper" different appearance. In another embodiment, the base coat 64 can include one type and amount of effect additive 74, while the effect coating 66 can include a different type and / or amount of effect additive 74 to produce an additional appearance. Thus, many variables affect the appearance of metallic colors, such as the type of flakes, the size of the flakes, the coating in which the flakes are included, the base color, etc. Thus, it is difficult to match metallic effects due to the various factors and appearances that are possible.

[0090] Pearlescent coatings include effect additives 74 that selectively reflect, absorb, and / or transmit visible light, which can produce a colored appearance that varies based on the structure and morphology of the flakes. This gives the coating a sparkle and depth that changes with viewing angle and / or illumination angle. The effect additives 74 in a pearlescent coating can be ceramic crystals, and these effect additives 74 can be added to the base coat 64, the effect coating 66, or both. Furthermore, the pearlescent effect additives 74 can be of varying grades with different sizes, refractive indices, shapes, etc., and different grades can be used individually or in combination. The pearlescent effect additives 74 can also be combined with metallic effect additives 74 in a variety of combinations, and the appearance of the coating will vary with changes in the concentration, type, location, etc. of the effect additives. All of the different possible variations in the effect additives 74 can be applied to a single color, so a technique to match only the color will be ineffective in reproducing the appearance of an effect pigment-based coating.

[0091] As noted above, the coating based on effect pigments has a varying appearance, and thus as Figure 3A , Figure 4A and Figure 5A illustrated, different pixels within the image will have different colors, brightness, hue, chroma, or other appearance characteristics. Thus, a color matching protocol that breaks the image into pixels and then analyzes the image pixel by pixel to determine pixel differences between two or more pixels can be helpful in matching the overall appearance of the coating based on effect pigments. The characteristics of the image can be determined based on the differences between the pixels within the image, with a mathematical model such as the machine learning model described above. As such, as noted above, the target image 58 can be analyzed pixel by pixel with the mathematical model to produce the target image characteristics 26, and this can be combined with other target image characteristics 26 such as color (which can be determined by looking at the target image 58 as a whole rather than pixel by pixel) or other target image characteristics. Again as noted above, the resulting target image characteristic(s) 26 can be compared to the sample database 30 of sample image characteristics 40 that have been produced for similar samples to find the best match. The pixel by pixel evaluation can produce a match for the effect pigments that would not be possible with an evaluation based on looking at the target image 58 as a whole.

[0092] Referring to the embodiment illustrated in Figure 12 and with continued reference to Figures 1-11 , as noted above, the imaging device 16 captures a target image 58 of the target coating 12. During the capture of the target image 58, the imaging device 16 is disposed at an imaging angle 76 relative to the surface of the target coating 12, and the illumination source 78 is disposed at an illumination angle 80. The target image 58 is divided into a plurality of target pixels 82, where the plurality of target pixels 82 varies such that at least one target pixel 82 is different in appearance from another target pixel 82.

[0093] In a similar manner, referring to the embodiment illustrated in Figure 13 and with continued reference to Figures 1-12 , again as noted above, the imaging device 16 captures a sample image 32 of the sample coating 60. The imaging device 16 used to capture the sample image 32 can be the same as the imaging device 16 used to capture the target image 58, but different imaging devices 16 can also be utilized. The illumination source 78 is also used to capture the sample image 32, where the illumination source 78 is similar to Figure 12As described above with respect to the target image 58, the imaging device is disposed at an imaging angle 76 relative to the surface of the sample coating 60, and the illumination source 78 is disposed at an illumination angle 80. The sample image 32 is divided into a plurality of sample pixels 84, one of which has a different appearance than another. In embodiments, the sample coating 60 includes an effect additive 74, which is shown in the illustration of the sample image 32 with small dots, and which produces a variation in the appearance of the sample pixels 84 within the sample image 32. Figure 12 and Figure 13 The target coating 12 and the sample coating 60 in

[0094] As shown in embodiments in Figure 14 and with continued reference to Figures 1-13 , a sample database 30 is generated for matching the target image 58 to the sample paint formulation 70. Generating the sample database 30 includes generating the sample paint formulation 70, where the sample paint formulation 70 includes the effect additive 74. The sample paint formulation 70 can include the base coat 64 and the clear coat 68, or the base coat 64, the effect coat 66, and the clear coat 68, but other embodiments are possible. For example, any of the base coat 64, the optional effect coat 66, and the clear coat 68 can include multiple layers, and other coatings can also be present.

[0095] Once the sample paint formulation 70 is generated, the sample coating 60 is generated using the sample paint formulation 70. This is typically done by applying the material from the sample paint formulation 70 to the substrate 14, for example, via spray, application with a brush, dip coating, digital printing, or any other coating technique. In embodiments, the sample coating 60 is formed by spray, where the spray utilizes the recommended spray conditions for the paint grade in the sample paint formulation 70. The spray conditions can include the type of solvent, the amount of solvent, the spray gun pressure, the type and / or size of the nozzle of the spray gun used, the distance between the spray gun and the substrate 14, and the like. Producing the sample coating 60 using the same techniques typically used by an automotive body shop or other situation that can need to match a target coating 12 can provide a more accurate representation of the finished product that can be expected than if the sample coating 60 is applied with another technique. The sample paint formulation 70 includes one or more effect additives 74, so the sample coating 60 has a varied appearance, where one portion of the sample coating 60 appears different than another portion. For example, a sparkle appears different than a matte color.

[0096] A sample image 32 is then generated using the imaging device 16, such as by taking a picture of the sample coating 60. The sample image 32 includes a plurality of sample image data 38, such as RGB values, L*a*b* values, etc. The sample image 32 can be one or more still images or moving images. In embodiments, the sample image 32 includes a plurality of still images captured with known and specified lighting, imaging angle, distance between the imaging device 16 and the sample coating 60, and lighting angle. The sample image 32 can also be for a substantially flat portion of the sample coating 60, but in some embodiments, the sample image can also include one or more images of the sample coating 60 having a known curvature.

[0097] The feature extraction analysis process 28" can be applied to the sample image data 38 to generate sample image features 40. The sample image features 40 are then associated with the sample paint formulation 70 and saved in the sample database 30. The sample database 30 can include a plurality of sample image features 40 associated with one sample paint formulation 70, and one or more sample images 32 can also be associated with the sample paint formulation 70. The sample database 30 is saved in one or more storage devices 22, which can be the same or different from the storage devices 22 used to match the target coating 12 described above. As described above, the storage devices 22 that save the sample database 30 can be associated with the data processor 24 to execute instructions for saving data and retrieving data from the sample database 30. The process shown in FIG. 4 can be repeated for a variety of sample paint formulations 70, so that a plurality of sample paint formulations 70 are saved in the sample database 30. The sample paint formulations 70 can also be configured to nearly match original equipment coatings provided by vehicle manufacturers, so that the vehicle coating matches the sample paint formulation 70. One or more sample images 32 can be utilized as the calculated matching sample image 44, or a different set of parameters can be utilized to capture the calculated matching sample image 44 from the sample coating 60. Figure 14 The feature extraction analysis process 28" can be applied to the sample image data 38 to generate sample image features 40. The sample image features 40 are then associated with the sample paint formulation 70 and saved in the sample database 30. The sample database 30 can include a plurality of sample image features 40 associated with one sample paint formulation 70, and one or more sample images 32 can also be associated with the sample paint formulation 70. The sample database 30 is saved in one or more storage devices 22, which can be the same or different from the storage devices 22 used to match the target coating 12 described above. As described above, the storage devices 22 that save the sample database 30 can be associated with the data processor 24 to execute instructions for saving data and retrieving data from the sample database 30. The process shown in FIG. 4 can be repeated for a variety of sample paint formulations 70, so that a plurality of sample paint formulations 70 are saved in the sample database 30. The sample paint formulations 70 can also be configured to nearly match original equipment coatings provided by vehicle manufacturers, so that the vehicle coating matches the sample paint formulation 70. One or more sample images 32 can be utilized as the calculated matching sample image 44, or a different set of parameters can be utilized to capture the calculated matching sample image 44 from the sample coating 60.

[0098] A plurality of sample paint formulations 70 can be generated that produce nearly identical sample images 32, where the sample paint formulations 70 differ from one another and include different grades of paint. Some vehicle repair shops tend to utilize one grade of paint, and the grade of paint can vary from one vehicle repair shop to the next. Because vehicle repair shops have a skill and familiarity with utilizing a particular grade of paint, a sample database 30 that includes matching sample paint formulations 70 that utilize the same grade of paint as the vehicle repair shop is familiar with can improve results. In this manner, different vehicle repair shops (or other users of the sample database 30) that typically use different grades of paint can utilize the same sample database 30.

[0099] exist Figure 15 In the implementation method and continue to refer to Figures 1-14 The above details are shown in more detail. Figure 14 The described feature extraction and analysis process 28” is used to divide the sample image 32 into a plurality of sample pixels 84. Each sample pixel 84 has sample pixel image data, and the sample pixel image data is used to determine the sample pixel features 86 of the sample pixel 84. Because the sample pixel 84 may have an appearance different from the overall sample image 32, the sample pixel features 86 may be different from the overall sample features 40. The sample pixel features 86 may be the determination of the RGB value of the sample pixel 84, the L*a*b* value of the sample pixel 84, or other appearance attributes of the sample pixel 84. In one embodiment, the sample pixel features 86 are determined for all sample pixels 84 of the sample image 32, but in an alternative embodiment, the sample pixel features 86 may be determined for only a subset of the sample pixels 84 of the sample image 32. In all embodiments, the sample pixel features 86 are determined for a plurality of sample pixels 84, wherein the sample pixel features 86 vary for at least some of the sample pixels 84. Then, sample pixel feature differences 88 are determined for the sample pixels 84. Then, sample image features 40 are determined based on the sample pixel feature differences 88.

[0100] exist Figure 16 China and continue to refer to Figures 1-15 An implementation utilizing a three-dimensional coordinate system 90 is illustrated. The three-dimensional coordinate system 90 can represent the RGB color system, wherein one axis of the three-dimensional coordinate system 90 is the R value, another axis is the G value, and a third axis is the B value. Alternatively, the three-dimensional coordinate system 90 can represent the L*a*b* color system, wherein one axis is the L* value, another axis is the a* value, and a third axis is the b* value. In alternative embodiments, the three-dimensional coordinate system 90 can represent other axes, and in some embodiments, the three-dimensional coordinate system 90 may not be utilized. Sample pixel features 86 can then be plotted in the three-dimensional coordinate system 90, and the number of sample pixels 84 falling within each block of the three-dimensional coordinate system 90 can be recorded. Figure 16 In the hypothetical illustration, the first block closest to the 0-0-0 coordinates has a count of 1, the block directly above the first block has a count of 2, and the block directly above the block with a count of 2 has a count of 5. Substitution techniques can be used to determine the sample pixel feature differences 88, and multiple techniques can be used to determine the multiple sample pixel feature differences 88 of a sample image 32 in multiple ways.

[0101] Then, as Figure 15 As shown and continue to refer to Figures 1-14 and Figure 16The sample image features 40 can be determined from the sample pixel feature differences 88. Because the sample image features 40 are based on color or appearance variations in different spaces within the sample image 32, determining the sample image features 40 from multiple sample pixel feature differences 88 is referred to herein as spatial micro color analysis. There can be multiple sample image features 40 determined for one single sample image 32, and some of those sample image features 40 can be based on spatial micro color analysis, while others can not be based on spatial micro color analysis. However, in this specification, at least one of the sample image features 40 is based on spatial micro color analysis. As such, the sample database 30 includes at least one sample image feature 40 that is based on spatial micro color analysis. Figure 15 the sample image data 38 block at the top of Figure 15 the sample image features 40 block at the bottom of Figure 14 embodiments of the steps utilized between the likewise labeled blocks in The computed match sample image 44 can also be saved in the sample database 30, where the computed match sample image 44 can be used to display to a user as described above. The user can then determine whether the computed match sample image 44 is an acceptable match for the target coating 12 based on visual analysis. The computed match sample image 44 can be captured with the imaging device 16 if it is acceptable for visual analysis.

[0102] Reference is made to the embodiments shown in Figure 17 and with continued reference to Figures 1-16 . Figure 17 Embodiments for utilizing spatial micro color analysis on the target image 58 are shown, as exemplified by the feature extraction analysis process 28’ block in Figure 2 . Thus, Figure 17 the target image data 18 block at the top of Figure 16 the third target image features 26 block from the bottom of Figure 2 are taken from the likewise named blocks in Figure 17 Embodiments are also shown where the mathematical model 100 used to compare the target image features 26 to the sample image features 40 is a model other than the machine learning model 42. However, it should be understood that the machine learning model 42 can still be utilized in some embodiments. As described above, Figure 2 and Figure 17 The sample image features 40 shown in

[0103] As described above, Figure 12 and Figure 17As shown, the target image 58 is divided into a plurality of target pixels 82. Each target pixel 82 includes target pixel image data, where the target pixel image data is at least a portion of the target image data 18. As described above for the sample pixel features 86, target pixel features 102 are determined from the target pixels 82, where each of the target pixels 82 has at least some of the target image data 18, and the at least some of the target image data 18 varies for different target pixels 82. As again described above for the sample pixel features 86, target pixel feature differences 104 are then determined from the target pixel features 102, and the target image features 26 are then determined from the target pixel feature differences 104. As such, at least one of the target image features 26 is determined by spatial micro color analysis because at least some of the target pixels 82 have different target image data 18. The target image features 26 for a single target image 58 can include one or more target image features 26 determined by spatial micro color analysis, and can also include one or more target image features 26 that are not based on spatial micro color analysis. As described above, the computed matching sample image 44 is then determined with the pre-specified matching criteria 46.

[0104] A variety of spatial micro color analysis mathematical techniques can be utilized to determine the target image features 26 and / or the sample image features 40, and in some embodiments the same techniques can be utilized to facilitate matching. A partial list of spatial micro color analysis mathematical techniques is listed below, where the general term pixel means either a target pixel 82 or a sample pixel 84, and the general term image means either a target image 58 or a sample image 32. Examples of spatial micro color analysis mathematical techniques include, but are not limited to: determining L*a*b* color coordinates of individual pixels of an image; determining average L*a*b* color coordinates of individual pixels from a full image of an image; determining flash regions of a black and white image of an image; determining flash intensity of a black and white image of an image; determining flash rank of a black and white image of an image; determining flash color of an image; determining flash clusters of an image; determining flash color differences within an image; determining flash persistence of an image, where flash persistence is a measure of flash as a function of one or more lighting changes during capture of the image; determining pixel level color constancy in the presence of one or more lighting changes during capture of an image; determining wavelet coefficients of an image at a pixel level; determining Fourier coefficients of a target image at a target pixel level; determining average color of a local region within an image, where a local region can be one or more pixels, but where a local region is less than a total region of an image; determining pixel count within a discrete L*a*b* range of an image, where the L*a*b* range can be fixed, or can be data driven such that the range varies; determining maximum fill coordinates of a cubic bin of an image at a pixel level, where a cubic bin is a 3-dimensional coordinate mapping based on using L*a*b* or RGB values; determining overall image color entropy of an image; determining image entropy of one or more planes in an L*a*b* plane as a function of a 3rd dimension of an image; determining image entropy of one or more planes in an RGB plane as a function of a 3rd dimension of an image; determining local pixel variation metric of an image; determining granularity of an image; determining a high variance vector of an image, where the high variance vector is established using principal component analysis; and determining a high kurtosis vector of an image, where the high kurtosis vector is established using independent component analysis.

[0105] Determining the L*a*b* color coordinates of individual pixels of a target / sample image includes breaking the image into pixels and then determining the L*a*b* color coordinates of a plurality (or in some embodiments all) of the pixels. Determining the average L*a*b* color coordinates of individual pixels from the full image of the image means taking the mean of the L*a*b* samples for each pixel within the image. Determining the glint area of the black and white image of the image means taking or obtaining the black and white image and then determining the number of pixels that include glints and the total number of pixels within a given area. The given area can be the entire image or a subset of the image. The glint area is then determined by dividing the number of pixels that include glints in the given area by the total number of pixels in the given area. Determining the glint intensity of the black and white image of the image means determining the average brightness of the pixels that include glints within a given area.

[0106] Determining the glint rank of the black and white image means determining a value that describes the visual perception of the glint phenomenon from the glint area and the glint intensity. Determining the glint color of the image means determining the location of the glint and then determining the color of the glint. Determining the glint clustering of the image means using a clustering or distribution fitting algorithm to determine the various colors of glints that exist in the image. As noted above, determining the glint color difference within the image means determining the color of the glint in the image and then determining the variation or difference in that color in the glint. Determining the glint persistence of the image means determining whether the glint remains within a given pixel (and whether the glint has a change in brightness) when the lighting changes during the capture of the image and the imaging angle 76 and the lighting angle 80 remain the same. Determining the pixel level color constancy means determining whether the color of a pixel within a given pixel remains the same when the lighting changes during the capture of the image and the imaging angle 76 and the lighting angle 80 remain the same. At least two different images are required to determine glint persistence and color constancy.

[0107] Determining pixel-level Fourier coefficients of an image means determining coefficients associated with the image after using a Fourier transform to decompose the image into sinusoidal components of different frequencies, where the coefficients describe the frequencies present in the image. This can be determined for a given number of pixels within a given region of the image. Determining pixel-level wavelet coefficients of an image means determining coefficients associated with the image after using a discrete or continuous wavelet transform to decompose the image into components associated with shifted and scaled versions of a wavelet, where the coefficients describe the image content associated with the relevant shifted and scaled versions of the wavelet. A wavelet is a function that tends to zero at extremes and contains oscillations in local regions. This can be determined for a given number of pixels within a given region of the image. Determining the average color of a local region within an image means determining the average color of one or more pixels within a sample region that is less than the total region of the image. Determining the pixel count within a discrete L*a*b* range of an image means determining the pixel count within a given region of a three-dimensional coordinate system 90 using L*a*b* values as axes, with Figure 16 similar to the illustration of FIG. 1. The size of the given region within the three-dimensional coordinate system 90 can be fixed, or the size can also vary depending on the generated count value. Determining the maximum filled coordinates of a pixel-level cubic bin of an image, where a cubic bin is based on a three-dimensional coordinate system 90 using L*a*b* or RGB values as axes, means determining the block with the highest pixel count as Figure 16 illustrated in FIG. 2. The value that designates a block as being "maximally" filled can be determined using a set value, a percentage value, or other metric.

[0108] Determining the overall image color entropy of an image means determining the overall color entropy of an image based on variations within pixels. The color entropy can be a Shannon entropy or other type of entropy calculation. Determining the image entropy in one or more planes of the L*a*b* plane or the RGB plane as a function of the 3rd dimension of the image means selecting one or more planes of the L*a*b* plane or the RGB plane as a function of the 3rd dimension of the image as Figure 16the plane to determine entropy. As noted above, the entropy can be a Shannon entropy or other type of entropy calculation. Determining a local pixel variation measure of the image means selecting substantially any pixel feature and then determining variation of that pixel feature within the image. Determining a granularity of the image means determining an impression of the granularity of the image based on shading or other features that imply granularity. Determining a high-variance vector of the image means determining a vector based on a vector origin at an origin of coordinates, where the high-variance vector is established using principal component analysis. The high-variance vector can be considered a target image feature 26 or sample image feature 40, or pixel data from the image can be projected onto the vector to result in a new target image feature 26 or sample image feature 40. Determining a high-kurtosis vector of the image also means determining a vector based on a vector origin at an origin of coordinates, where the high-kurtosis vector is established using independent component analysis. The high-kurtosis vector can be considered a target image feature 26 or sample image feature 40, or pixel data from the image can be projected onto the vector to result in a new target image feature 26 or sample image feature 40.

[0109] As Figure 18 illustrated in FIG. 6 and with continued reference to Figures 1-17 In another embodiment, a method of matching a target coating 12 is provided. The method includes obtaining 1200 a target image 58 of a target coating 12, where the target coating 12 is an effect pigment-based coating that includes an effect additive 74. The method also includes applying 1210 a feature extraction analysis process 28' to the target image 58, where the feature extraction analysis process 28' includes dividing 1220 the target image 58 into a plurality of target pixels 82 that include target pixel image data, determining 1230 target pixel features 102 for each target pixel of the plurality of target pixels 82, determining 1240 target pixel feature differences 104 between each target pixel 82, determining 1250 target image features 26 from the target pixel feature differences 104, and calculating 1260 a calculated matching sample image 44 using the target image features 26 based on substantially satisfying one or more pre-designated matching criteria.

[0110] As Figure 19 illustrated in FIG. 6 and with continued reference to Figures 1-18A method of generating the sample database 30 is also provided. The method includes preparing 1300 a sample coating 60 from a sample paint formulation 70 that includes an effect additive 74 such that the appearance of the sample coating 60 varies from one location to another. The method also includes imaging 1310 the sample coating to generate a sample image 32 that includes sample image data, where the sample image 32 is divided into a plurality of sample pixels 84, each sample pixel including sample pixel image data. The next step is retrieving 1320 one or more sample image features 40 from the sample image data, where at least one of the sample image features 40 includes a spatial micro color analysis that includes a value determined from a sample pixel feature difference 88 between at least two of the sample pixels 84. Another step is saving 1330 the sample paint formulation 70 and the one or more sample image features 40 in the sample database 30, where the sample paint formulation 70 is associated with the one or more sample image features 40.

[0111] Referring to Figure 20 Once the computed matching sample image 44 is obtained, it can be approved by an operator. At this point, a repair paint 110 can be prepared using the sample paint formulation 70 that corresponds to the computed matching sample image 44. In an example embodiment, the sample paint formulation 70 includes the effect additive 74 and one or more other components 72. The repair paint 110 can be applied to a substrate 14 such as a vehicle in need of repair using one or more of a variety of techniques. In an example embodiment, the repair paint 110 is applied to the substrate 12 with a digital printer 112 and digital printing techniques as shown, but in alternative embodiments, the repair paint 110 can be applied to the substrate 12 by other techniques including but not limited to spraying, application with a brush, and / or dipping.

[0112] Figure 21 A method for analyzing one or more features of a painted automotive component or an entire vehicle is depicted at 1400. Examples of types of analysis include determining the type of paint chip used on a painted object, and using the analysis for matching purposes. The paint chip selection process can be, for example, an automated part of a paint matching formulation development workflow for effect paints. This is in contrast to previous methods that use a manual process to identify possible effect paints for matching purposes.

[0113] Image data 1404 of the brushed object is provided to an analysis process. An electronic imaging device with intrinsic or supplemental illumination 1402 can generate image data 1404 of the target coating of the brushed object for use in one analysis computer program or multiple analysis computer programs. The image data 1404 is placed into a particular color space, such as L*, a*, b*, at 1406 so that pre-designated features can be extracted. The extracted features can include any color or appearance features needed for the matching process.

[0114] Feature extraction process 1408 parses the image data into regions of sub-images 1410. The degree of presence (or absence) of features within the sub-images is used to generate a distribution 1414 of features across the sub-images 1410 at 1412. In addition, statistical measures are generated at 1412 to more fully describe the distribution 1414 of features within the image data 1404. For example, statistical measures used to describe the distribution 1414 of features can include standard deviation, mean, kurtosis measure, skew measure, etc. Note that the spatially localized color phenomena in metallic paints cannot be captured by reflectance alone.

[0115] A computer-implemented non-transitory storage medium 1416 or multiple non-transitory storage media 1416 stores the distribution of extracted features and statistical data. Machine learning models 1418 are applied to the stored data to predict the type of one or more flakes present in the target coating. The machine learning models 1418 can employ supervised training, unsupervised training, or a combination thereof. Examples of suitable machine learning models include, but are not limited to: linear regression, decision trees, k-means clustering, principal component analysis (PCA), random decision forests, neural networks, or any other type of machine learning algorithm known in the art. In an implementation, the machine learning models are based on a multi-layer artificial neural network.

[0116] The one or more machine learning models 1418 predict the type of flakes present in the target coating. The prediction output 1420 can represent the ratio of the most likely type of flakes present in the target coating to the least likely type of flakes present. The formulation process can use the automatically determined type of flakes as part of the paint formulation used to match the target coating.

[0117] Figure 22 An example configuration of the processes in 1500 is depicted at 1502. Reference is made to Figure 21 Figure 22 ​The illumination source 1402 can be an electronic imaging device for capturing images at a wide range of electromagnetic wavelengths including visible wavelengths or non-visible wavelengths. The electronic imaging device can include one or more different types of cameras 1502, such as a camera that is part of a mobile device. Examples of mobile devices include, but are not limited to, a mobile phone (e.g., a smartphone), a mobile computer (e.g., a tablet or laptop computer), a wearable device (e.g., a smartwatch or earpiece), or any other type of device known in the art that is configured to receive target image data. The illumination of the coated target can be at different angles to capture different patch sizes and geometries.

[0118] After capturing the image data 1402, the system can represent the color space using a plurality of matrices (e.g., three matrices, etc.). For example, a data structure for storing three matrices 1504 can be configured in the following manner. In implementations, three separate data structures can store each of an L* value matrix, an a* value matrix, and a b* value matrix. These values are for each pixel in the entire captured image or a portion of the captured image. The values stored in the matrices are converted from an RGB color space to an L*, a*, b* color space. The RGB color space is device dependent (e.g., different for each camera). The conversion to the L*, a*, b* color space makes subsequent calculations device independent.

[0119] Based on the color-related data stored in the data structures, a feature extraction process 1408 selects which features to analyze for determining one or more patch types of the target coating. For example, the feature extraction process can extract features of the sub-image 1410 at 1506, such as average L*, a*, b* color coordinates, sparkle intensity, patch size, etc.

[0120] Figure 23 An example configuration for further processing the data contained in the sub-image 1410 is depicted at 1600. The sub-image 1410 can be analyzed at 1602 to determine a distribution 1604 of one or more features across the sub-image. For example, the system can determine how many pixels within the sub-image include a sparkle associated with a patch and a total number of such pixels. The system can further analyze the “size” feature associated with the patch to determine a distribution 1604 of patch sizes across the sub-image 1410.

[0121] Figure 24 An example configuration of a machine learning model that is applied to the feature distributions and statistics stored in one or more color data storage media 1416 at 1702 to generate predicted probabilities 1704 is depicted at 1700. The example configuration uses an artificial neural network 1706 to predict patch types based on input feature distributions and statistics.

[0122] The artificial neural network 1706 can be comprised of an input layer 1708, one or more hidden layers 1710, and an output layer 1712. The input layer 1708 can be provided with one or more input nodes to receive the distribution of features and statistical information related to the image from the target coating. The one or more hidden layers 1710 within the artificial neural network 1706 help identify interrelationships within the input data. Because the interrelationships can be non-linear in nature, the artificial neural network 1706 can better identify patterns or desired objects (e.g., patches or pixels associated with patches) within the data set from one color data storage medium 1416 or multiple color data storage media 1416. The nodes of the artificial neural network 1706 can use many different types of activation functions, for example, sigmoid activation functions, logistic activation functions, and the like.

[0123] The output layer 1712 of the artificial neural network 1706 provides a prediction at 1714 indicating the probability that a particular patch is present. The prediction output can represent the ratio of the most likely type of patch present in the target coating to the least likely type of patch. The prediction can also be provided as the top n set of patch results.

[0124] In one embodiment, the artificial neural network executing on one or more processors is a convolutional neural network trained on a labeled image data set to learn features of each type of object within the image. The labeled image data set includes information identifying the labeled sub-image and the class of each object. During training, the artificial neural model can predict the class based on the features. Because the interrelationships between the training input images and the classes are predefined, the training module can adjust the model’s predictions to match the predefined associations between the input images and the features. Once the model has been trained, the machine learning classifier can predict the patch class based on the data.

[0125] It should be appreciated that other machine learning models can be used, for example, random forest models, decision tree models. The machine learning models used within the systems disclosed herein include empirical machine learning models trained on data generated during actual applications (e.g., color related data resulting from a spray application, etc.).

[0126] It should also be appreciated that the output prediction probabilities from the model can be further analyzed. Figure 25The following example is provided at 1800, where the color analysis expert system 1804 applies its color analysis rules 1808 to the output prediction probabilities 1704 to determine whether one or more output candidate patches are more or less likely to be present in the sub-image of the target coating. For example, the color analysis expert system 1804 can determine that two patches are not typically present for the coating, and adjust the probability predictions at 1806 to reflect that actual level of detail.

[0127] Figure 26 The system at 1900 helps a colorist understand and correct for spatial color differences by analyzing and correcting for differences in multi-angle spatial average colors. This is beneficial for tools such as image-based color formula retrieval systems used by colorists.

[0128] Figure 26 The system receives a target image of a target coating containing target image data. The color model 1902 and the local color model 1906 help predict a color difference 1914 between the target coating and a sample coating based at least in part on average L*, a*, b* 1904 generated by the color model 1902 and color population 1908 generated from the local color model 1906.

[0129] As this information is provided by the color model 1902, the local color model 1906 predicts spatial color features (e.g., color entropy features, etc.) from image data that are not “average” data. To generate the color population 1908, the color space is divided into primary, secondary, and tertiary color regions as shown at 1910. As a non-limiting example, each pixel in the image can be classified based on the color system to which each pixel belongs in the image. Another non-limiting approach includes modeling the distribution of colors and determining which group the color belongs to. Such an approach provides an overall representative classification of the color population. In embodiments, the local color model generates the color population by, for example, forming groups or clusters of similar color feature data. In embodiments, a k-means clustering method can form clusters of primary L*, a*, b* values, clusters of secondary L*, a*, b* values, and clusters of tertiary L*, a*, b* values. In this example, “k” in the k-means clustering method would be 3, such that three clusters are formed.

[0130] Predicted color difference 1914 (e.g., delta E) is generated by optimization routine 1912 using output from color model 1902 and local color model 1906. Optimization routine 1912 can include, for example, a numerical algorithm to generate a formulation that optimizes the predicted color difference from the standard. In implementations, the numerical algorithm tests different formulations and determines which formulation has the smallest predicted color difference from the standard. Non-limiting examples include a function that generates a color difference metric based on the paint formulation, a model of the color phenomenon, and a color metric associated with the standard. The formulation that minimizes the function is selected. In this way, empirical models that involve the formulation and various localized color features can be used for color matching.

[0131] Different optimization routines for the function can be used, e.g., nonlinear programming numerical methods, linear programming numerical methods, etc. In this optimization routine, the function can be minimized to identify the smallest predicted color difference from the standard. The function can generate, for example, a prediction of the average delta E of color model 1902 and local color model 1906 from the standard.

[0132] In implementations, a relaxed constraint process is used to optimize the predicted color difference 1914. The relaxed constraint process can include optimizing a target function of color difference subject to constraints whose coefficients are allowed to be relaxed. A penalty function is used within the color difference optimization routine to provide constraint relaxation. The optimized color difference determines the automotive paint components for spraying the substrate at 1918. The model prediction 1916 can be sprayed with an acceptable delta E, such that an actual delta E 1920 can be obtained. The sprayed sample 1922 is then compared to the standard 1924 to determine whether an acceptable match has been obtained (e.g., delta E between the sample and the standard is within a pre-specified threshold, etc.). If the match is not acceptable, the process can iterate until an acceptable match is achieved, as shown in Figure 27

[0133] Figure 27 A process flow for determining when a colorizer has achieved an acceptable match is depicted at 2000. The colorizer performs an initial spray at 2002 to attempt to match the target coating. Color space data is obtained by a color measurement device (e.g., spectrophotometer, etc.).

[0134] The color difference 2004 between the initial spray and the target coating is provided as input at 2006 to the color model with bias and at 2008 to the local color model with predicted bias. The predicted bias can be determined based on the difference between the predicted delta E and the actual delta E obtained after the selected formulation is used to spray the sample. This bias is used to correct the prediction after the first iteration.

[0135] ​The two color models can be optimized in many different ways at 2010. In an embodiment, the optimization can be performed using a relaxation constraint method. The optimization results in new predicted color differences 2012. The colorant is used in the paint formulation software to generate a new formulation to be sprayed at 2014 using the predicted color differences 2012. If the color differences 2016 from the additional spray are acceptable as determined at 2018, then the colorant process is complete at 2020. If the color differences are not acceptable as determined at 2018, then additional spray and analysis iterations are performed until acceptable color differences are obtained as determined at 2018.

[0136] Figure 28 The system is depicted at 2100 that is extended to analyze the target coating for texture matching. Texture matching can include any feature that describes a local color phenomenon. It should be noted that this does not include orange peel caused by sprayed paint.

[0137] In another embodiment, texture matching is technically difficult because the texture can vary over a given area, for example not only in terms of granularity but also in terms of glossiness, micro-luster, haze, mottling, speckle, sparkle, or shimmer. Depending on the size and texture of the small building blocks of the material, the texture to be matched can be described as a visual surface structure in the plane of the paint film. Texture does not include paint film roughness such as orange peel, but includes visual irregularities in the plane of the paint film. Structures smaller than what the human eye can see contribute to color, while larger structures contribute to texture.

[0138] The system checks the color population of the target coating, the texture features, and the delta "E" (color difference) across the target at 1920 to determine the granularity distribution across the target coating. The texture features are extracted at 1918 for matching the texture properties of the paint film (sprayed at 1916) on the substrate to be matched.

[0139] It should be understood that the texture machine learning model can calculate a paint formulation with matching texture properties. The machine learning model can examine the particle size distribution of the paint to be matched, for example by examining dark and light areas in the image data. A machine learning model such as a convolutional neural network can be applied to the texture dataset to predict a formulation based on the delta E measurements while not deviating from acceptable color matching.

[0140] In an implementation, the convolutional neural network can be configured such that its convolution operators extract texture features from the input image. The configuration is built through supervised training such that the configuration preserves the spatial relationships between pixels in the image. More specifically, the convolutional neural network learns optimal filter values, where the number of filters is set to extract a desired number of texture features. For example, if the texture matching process requires more texture features, the number of filters will be increased accordingly.

[0141] Figure 29 A server environment in which a user 2202 can interact with a color and appearance matching analysis system 2204 is depicted at 2200. The user 2202 can interact with the system 2202 in many ways, such as through one or more networks 2206. A server 2208, accessible through the network 2206, can host the system 2204. One or more data stores 2210 can store color analysis data 2212 to be analyzed by the system 2204, as well as any intermediate or final data generated by the system.

[0142] Figure 29 The system in the foregoing detailed description can be a web-based integrated reporting and analysis tool that provides a user with flexibility and functionality for performing color and appearance matching analysis. It should be appreciated that the system can also be set up on a standalone computer for access by a user.

[0143] While at least one implementation has been presented in the foregoing detailed description, it should be appreciated that a vast number of modifications can be made to the implementations described without departing from the scope disclosed herein. It should also be appreciated that the implementations are only examples and are not intended to limit the scope, applicability, or configuration in any way. Rather, the foregoing detailed description will provide those ordinarily skilled in the art with a convenient road map for implementing an implementation, but it should be appreciated that various changes can be made without departing from the scope as set forth in the claims and their equivalents.

Claims

1. A system for matching a target coating, comprising: Storage device, used to store instructions; One or more data processors are configured to execute instructions to: Receive a target image of the target coating, wherein the target image includes target image data; The target image data is subjected to feature extraction and analysis processing to determine the target image features, wherein the feature extraction and analysis processing includes dividing the target image into sub-images containing multiple target pixels, wherein the sub-images may include a single patch or not include a patch; Determine the target pixel features of the sub-image, wherein determining the target pixel features of the sub-image includes determining the distribution of the target pixel features across the sub-image, and A machine learning model is applied to identify one or more types of patches present in the target coating using the determined distribution of the target pixel features across the sub-image.

2. The system according to claim 1, wherein, Determining the target pixel features of the sub-image includes generating cumulative distribution statistics based on predetermined color attributes.

3. The system according to claim 1, wherein, The one or more data processors are configured to apply the feature extraction analysis process, wherein the feature extraction analysis process determines one or more of the L*a*b* color coordinates of each target pixel of the target image and the overall image color entropy of the target image.

4. The system according to claim 1, wherein, The one or more data processors are configured to execute instructions to retrieve a mathematical model for determining a computed matching sample image.

5. The system according to claim 4, wherein, The mathematical model is a machine learning model.

6. The system according to claim 1, wherein, The one or more data processors are configured to determine the paint formulation corresponding to the calculated matching sample image.

7. The system according to claim 6, wherein, The one or more data processors are configured to determine multiple paint formulations corresponding to the calculated matching sample images, wherein the multiple paint formulations include different grades of paint.

8. The system according to claim 1, wherein, The one or more data processors are configured to receive target image data of the target coating, wherein the target image data is associated with multiple images of the target coating having different light angles relative to the imaging device.

9. The system according to claim 1, wherein, The one or more data processors are configured to receive the target image data of the target coating, wherein the target image data is associated with multiple images of the target coating with different magnifications.

10. The system according to claim 1, wherein, The target coating is a metallic coating, a pearlescent coating, or a combination thereof.

11. A method for matching a target coating, comprising: A target image of the target coating is received by one or more data processors, wherein the target image includes target image data; The target image data is subjected to feature extraction and analysis processing by one or more data processors to determine target image features, wherein the feature extraction and analysis processing includes dividing the target image into sub-images containing multiple target pixels, wherein the sub-images may include a single patch or not include a patch; The target pixel features of the sub-image are determined by the one or more data processors, wherein determining the target pixel features of the sub-image includes determining the distribution of the target pixel features across the sub-image, and Machine learning models are applied through one or more data processors to identify one or more types of patches present in the target coating using the determined distribution of the target pixel features across the sub-image.

12. The method according to claim 11, wherein, Determining the target pixel features of the sub-image includes generating cumulative distribution statistics based on predetermined color attributes.

13. The method according to claim 11, wherein, The one or more data processors are configured to apply the feature extraction analysis process, wherein the feature extraction analysis process determines one or more of the L*a*b* color coordinates of each target pixel of the target image and the overall image color entropy of the target image.

14. The method according to claim 11, wherein, The one or more data processors are configured to execute instructions to retrieve a mathematical model for determining a computed matching sample image.

15. The method according to claim 14, wherein, The mathematical model is a machine learning model.

16. The method according to claim 11, wherein, The one or more data processors are configured to determine the paint formulation corresponding to the calculated matching sample image.

17. The method according to claim 16, wherein, The one or more data processors are configured to determine multiple paint formulations corresponding to the calculated matching sample images, wherein the multiple paint formulations include different grades of paint.

18. The method according to claim 11, wherein, The one or more data processors are configured to receive target image data of the target coating, wherein the target image data is associated with multiple images of the target coating having different light angles relative to the imaging device.

19. The method according to claim 11, wherein, The one or more data processors are configured to receive the target image data of the target coating, wherein the target image data is associated with multiple images of the target coating with different magnifications.

20. The method according to claim 11, wherein, The target coating is a metallic coating, a pearlescent coating, or a combination thereof.

Images