# US 2014/0195468 A1 · Method for matching sparkle appearance of coatings > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2014/0195468 A1 - Document ID: US20140195468A1 - Status: Published application - Kind: A1 - Application number: 14/132,389 - Filed: 2013-12-18 - Published: 2014-07-10 - Earliest claimed: 2011-06-20 - Assignees: Axalta Coating Systems IP Co., LLC - Applicants: Axalta Coating Systems IP Co., LLC - Inventors: Mahnaz MOHAMMADI; Larry Eugene STEENHOEK - Web page: https://patentgrep.com/patent/US20140195468A1 ## CPC - G06N 3/02: Neural networks - G06N 3/0499: Feedforward networks - G06N 3/09: Supervised learning - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract This disclosure is directed to a process for producing one or more predicted target sparkle values of a target coating composition. An artificial neural network can be used in the process. The process disclosed herein can be used for color and appearance matching in the coating industry including vehicle original equipment manufacturing (OEM) coatings and refinish coatings. A system for producing one or more predicted target sparkle values of a target coating composition is also disclosed. ## Claims ### Claim at position 1 -19. (canceled) ### Claim 20 A system for producing one or more predicted target sparkle values of a target coating composition, said system comprising: (i) a computing device; (ii) an artificial neural network computing program product residing on said computing device or a host computer functionally coupled to said computing device, said artificial neural network computing program product comprises: a) an input layer having one or more input nodes for receiving inputs; b) at least one hidden layer having one or more hidden nodes; c) an output layer having one or more output nodes for producing outputs; (iii) an input device and an output device functionally coupled to said computing device; and (iv) a computer program product comprising computing codes functionally coupled to said computing device for performing a computing process, said computing process comprises the steps of: (1) receiving training data of a plurality of training coatings into said one or more input nodes of said artificial neural network computing program product, said training data comprise measured color characteristics, measured sparkle values and an individual training coating formulation associated with each of the training coatings, and (2) training said artificial neural network computing program product with said training data to produce a trained artificial neural network that is capable of producing a predicted sparkle value from said output nodes based on a coating formulation and color characteristics associated with said coating formulation, wherein said trained artificial neural network is trained based on said measured color characteristics, said measured sparkle values and said individual training coating formulation associated with each of the training coatings. ### Claim 21 (depends on 20) The system of claim 20, wherein said trained artificial neural network comprises individual trained input connection weights connecting each of the input nodes with each of the hidden nodes in a first of said hidden layers; individual trained output connection weights connecting each of the output nodes with each of the hidden nodes in a last of said hidden layers; and wherein said first and the last of the hidden layers are the same when there is only one hidden layer and said first and the last of the hidden layers are different when there are two or more hidden layers. ### Claim 22 (depends on 20) The system of claim 20, wherein said computing process further comprises the steps of: (3) receiving the target coating formulation and target color characteristics associated with said target coating formulation into said trained artificial neural network; and (4) producing the target sparkle value from said trained artificial neural network based on said target coating formulation and said target color characteristics associated with said target coating formulation; and (5) outputting said target sparkle value to said output device. ### Claim 23 (depends on 22) The system of claim 22, wherein target coating formulation and said target color characteristics associated with said target coating are received into said trained artificial neural network from said input device, from a target database functionally coupled to said computing device, or a combination thereof. ### Claim 24 (depends on 22) The system of claim 22 further comprising a display device selected from a high dynamic range (HDR) image display device, a non-HDR image display device, or a combination thereof. ### Claim 25 (depends on 24) The system of claim 24, wherein said computing process further comprising the steps of: (6) generating a target image having R,G,B values based on said predicted target sparkle values, said target color characteristics, and optionally, a shape of an article coated with the target coating composition; and (7) displaying said target image having said R,G,B values on said display device. ### Claim 26 (depends on 25) The system of claim 25 further comprising a BRDF computing program product functionally coupled to said computing device for producing a bidirectional reflectance distribution function (BRDF) based on said predicted target sparkle values, said target coating formulation, said target color characteristics associated with said target coating formulation, and a refractive index associated with said target coating formulation. ### Claim 27 (depends on 26) The system of claim 26, wherein said target image is generated as a HDR target image. ### Claim 28 (depends on 27) The system of claim 27, wherein said HDR target image is displayed on said HDR image display device, said non-HDR image display device, or a combination thereof. ### Claim 29 (depends on 25) The system of claim 25, wherein said target image is displayed based on one or more illumination angles, one or more viewing angles, or a combination thereof. ### Claim 30 (depends on 20) The system of claim 20, wherein said training data are received into said artificial neural network from said input device, from a training database functionally coupled to said computing device, or a combination thereof. ### Claim 31 (depends on 20) The system of claim 20 further comprising a high dynamic range (HDR) or wide dynamic range (WDR) imaging device.