Systems and methods for image based component surface analysis

Abstract

A system for assessing a surface condition of a component of an aftertreatment system is provided. The system includes a computing system coupled to an image capture device. The computing system is configured to: receive, from the image capture device, a first image of a surface of the component; extract a first feature from the first image; identify a second feature that corresponds to the first feature based on comparing the first feature to a set of historical features comprising at least the second feature; responsive to identifying the second feature, receive a condition of the component that corresponds to the second feature; and generate a notification for display on a user interface indicating the condition of the component.

Description

Atty. Dkt. No.: 106389-9772SYSTEMS AND METHODS FOR IMAGE BASED COMPONENT SURFACE ANALYSIS CROSS-REFERENCE TO RELATED APPLICATION[00011 The present application claims the benefit of and priority to Indian Provisional Application No. 202541029716, filed March 28, 2025, titled “SYSTEMS AND METHODS FOR IMAGE BASED COMPONENT SURFACE ANALYSIS,” which is incorporated herein by reference in its entirety and for all purposes.TECHNICAL FIELD

[0002] The present disclosure relates to diagnostics for systems, and particularly vehicle, components. In particular, the present disclosure relates to systems and methods for imagebased analysis of a component(s) of an exhaust aftertreatment system for diagnostic and / or servicing.BACKGROUND

[0003] Many engines are coupled to exhaust aftertreatment systems that reduce harmful exhaust gas emissions (e.g., nitrous oxides (NOx), sulfur oxides, particulate matter, etc.). The aftertreatment system may include one or more components, such as a filter, to remove harmful particles from exhaust. Over time and usage, the efficacy of these components may decrease. It may be desirable to identify potential failure modes for various components to keep the aftertreatment system operating as desired.SUMMARY

[0004] One embodiment relates to a system for assessing a surface condition of a component of an aftertreatment system. The system includes a computing system coupled to an image capture device. The computing system includes one or more processors and one or more memory devices storing instructions therein that, when executed by the one or more processors, cause the one or more processors to perform operations comprising: receiving, from the image capture device, a first image of a surface of the component; extracting a first feature from the first image; identifying a second feature that corresponds to the first featureAtty. Dkt. No.: 106389-9772based on comparing the first feature to a set of historical features comprising at least the second feature; responsive to identifying the second feature, receiving a condition of the component, wherein the condition of the component corresponds to the second feature; and generating a notification for display on a user interface, the notification indicating the condition of the component.

[0005] Another embodiment relates to a method for assessing a surface condition of a component of an aftertreatment system. The method includes: receiving, by a computing system, a first image of a surface of the component of the aftertreatment system; extracting, by the computing system, visual data from the first image; classifying, by the computing system, the visual data based on at least one of a distribution pattern or a color contained in the visual data; generating, by the computing system, a report regarding a condition of the surface of the component; generating, by the computing system, a recommendation regarding maintenance of the aftertreatment system; and providing, by the computing system, a maintenance recommendation to a user device, the maintenance recommendation comprising a notification indicating the condition of the component.

[0006] Yet another embodiment relates to a system for generating maintenance recommendations based on images of an aftertreatment system. The system includes a computing system including one or more processors and one or more memory devices storing instructions therein that, when executed by the one or more processors, cause the one or more processors to perform operations comprising: receiving an image of a component of the aftertreatment system; extracting visual data from the image of the component of the aftertreatment system; providing the visual data to a machine learning model configured to identify and classify information in the visual data; classifying, via the machine learning model, the visual data based on at least one of a distribution contained in the visual data, a color contained in the visual data, a pattern in the visual data, or an abnormality in the visual data, wherein the classification of the visual data includes comparing the visual data to stored historical visual data; characterizing, by the machine learning model, the image of the component based on the classification of the visual data; determining, by the machine learning model, a condition of the component; generating a report regarding the condition of the component of the aftertreatment system based on the determined condition; andAtty. Dkt. No.: 106389-9772outputting one or more maintenance recommendations regarding the component of the aftertreatment system to a display device.[00071 Numerous specific details are provided to impart a thorough understanding of embodiments of the subject matter of the present disclosure. The described features of the subject matter of the present disclosure may be combined in any suitable manner in one or more embodiments and / or implementations. In this regard, one or more features of an aspect of the invention may be combined with one or more features of a different aspect of the invention. Moreover, additional features may be recognized in certain embodiments and / or implementations that may not be present in all embodiments or implementations.BRIEF DESCRIPTION OF THE DRAWINGS

[0008] FIG. l is a schematic view of a block diagram of a system, according to an example embodiment.10009] FIG. 2 is a schematic view of a block diagram of an image recognition system used to assess an exhaust aftertreatment system, or a component(s) thereof, of the system of FIG. 1, according to an example embodiment.

[0010] FIG. 3 is a flow diagram of a method of assessing the aftertreatment system of the system of FIG. 1, according to an example embodiment.

[0011] FIG. 4 is a flow diagram of a method used to train a portion of a computing system used to assess the aftertreatment system of the system of FIG. 1, according to an example embodiment.

[0012] FIG. 5 is a view of a clean surface of a filter element, such as a diesel particulate filter of the system of FIG. 1, according to an example embodiment.

[0013] FIG. 6 is a view of a soot distribution on a surface of a filter element, particularly a diesel particulate filter, of the system of FIG. 1, according to an example embodiment.Atty. Dkt. No.: 106389-9772[00141 FIG. 7 is a view of a foreign contaminant accumulation on a surface of a filter element, particularly a diesel particulate filter, of the system of FIG. 1, according to an example embodiment.

[0015] FIG. 8 is a view of a locally concentrated soot accumulation on a surface of a filter element, particularly diesel particulate filter, of the system of FIG. 1, according to an example embodiment.DETAILED DESCRIPTION

[0016] Following below are more detailed descriptions of various concepts related to, and implementations of methods, apparatuses, computer-readable media, and systems for selectively facilitating assessing a component or components of an exhaust aftertreatment system. Before turning to the figures, which illustrate certain exemplary embodiments in detail, it should be understood that the present disclosure is not limited to the details or methodology set forth in the description or illustrated in the figures. It should also be understood that the terminology used herein is for the purpose of description only and should not be regarded as limiting.

[0017] As described herein, a system may include an engine and an exhaust aftertreatment system in exhaust gas receiving communication with the engine. The system may be a part of a vehicle (e.g., an on-road vehicle such as a truck or sedan or an off-road vehicle). The engine may be an internal combustion engine (ICE) configured to combust fuel. The aftertreatment system may include one or more components, such as a filter element (e.g., a diesel particulate filter (DPF)) configured to remove particulate matter, a dosing module (e.g., a doser) configured to supply a dosing fluid to the exhaust gas flowing in the exhaust gas system, and one or more catalyst devices configured to facilitate conversion of the exhaust gas constituents (e.g., nitrogen oxides, NOx) to less harmful elements (e.g., water, nitrogen), such as an oxidation catalyst (e.g., a diesel oxidation catalyst (DOC)), a selectively catalytic reduction (SCR) system, a three-way catalyst, and so on.

[0018] During operation of the aftertreatment system, various surfaces of the aftertreatment system, such as the filter element (e.g., DPF), the catalyst (e.g., DOC and / or the SCR), etc., may accumulate various kinds of deposits from soot and / or other harmful particles containedAtty. Dkt. No.: 106389-9772in the unfiltered exhaust gas. The accumulation of deposits and buildup of matter on these surfaces can cause deficiencies in their function, limiting and / or causing failures in their operation. When components of the aftertreatment system are deficient in their function, the cause for failure and the way in which the component is failing may be difficult and time consuming to diagnose manually (e.g., by a person).

[0019] Technically and beneficially, the systems, computer-readable media, and methods described herein can use image recognition to identify this accumulation of particles and / or deposits as a potential issue, identify a failure type, identify and / or predict a future failure, suggest potential causes for the failure, and / or suggest diagnostics to revert the failure or other recovery methods. In particular, the systems, computer-readable media, and methods described herein and, among other benefits, provide a technical solution to the technical problem of quickly and accurately diagnosing the condition of components of aftertreatment systems, and particularly surfaces of the components of aftertreatment systems. The systems, computer-readable media, and methods described herein provide a technical solution including receiving image data of one or more surfaces of an aftertreatment system and analyzing the image data to diagnose the condition of the surfaces of the aftertreatment system. Advantageously, the diagnosis process of the aftertreatment system may result in relatively fast, accurate diagnoses of components of the aftertreatment system while eliminating the need for manual inspection. These and other features and benefits are described more fully herein below.

[0020] Now referring to FIG. 1, a schematic view of a block diagram of a system 100 is shown, according to an example embodiment. The system 100 includes an engine 101 and an aftertreatment system 120 in exhaust gas receiving communication with the engine 101. In some embodiments, the system 100 may also include an image capture device 140 (as shown in FIG. 2), a controller 150, and a user device 240 (also shown in FIG. 2). In the configuration of FIG. 1, the system 100 is included in a vehicle. The vehicle may be any type of on-road or off-road vehicle including, but not limited to, wheel-loaders, fork-lift trucks, line-haul trucks, mid-range trucks (e.g., pick-up truck, etc.), sedans, coupes, tanks, airplanes, boats, and any other type of vehicle. In another embodiment, the system 100 may beAtty. Dkt. No.: 106389-9772embodied in a stationary piece of equipment, such as a power generator or genset. All such variations are intended to fall within the scope of the present disclosure.[00211 The engine 101 may be any type of internal combustion engine that generates exhaust gas, such as a gasoline, natural gas, or diesel fueled engine, and / or any other suitable engine. In the example depicted, the engine 101 is a part of a diesel engine system. In other embodiments, the engine 101 is part of a hybrid engine system having a combination of an internal combustion engine and at least one electric motor coupled to at least one battery. In some embodiments, the hybrid engine system may be configured as a mild-hybrid powertrain, a parallel hybrid powertrain, a series hybrid powertrain, or a series-parallel powertrain.

[0022] As shown in FIG. 1, an intake air throttle (IAT) valve 102, a fuel module or system 103, and an oil or lubrication system 104 are coupled to the engine 101. The IAT valve 102 is structured to control an amount of air supplied to the engine 101. The fuel module 103 is structured to provide fuel to the engine 101 (e.g., from a fuel source). The fuel module 103 may control one or more fueling parameters including a fuel amount, a fuel pressure, a fuel injection timing, etc. The oil system 104 is configured to provide a lubricant (e.g., lubricant oil) to the engine 101.

[0023] The aftertreatment system 120 is in exhaust-gas receiving communication with the engine 101. In the example depicted, the aftertreatment system includes a first catalyst member, shown as a diesel oxidation catalyst (DOC) 121, a filter (e.g., a particulate filter), shown as a diesel particulate filter (DPF) 122, and a second catalyst member, shown as a selective catalytic reduction (SCR) system 123. In some embodiments, the aftertreatment system 120 includes a third catalyst member, shown as an ammonia slip catalyst (ASC) 128. The DOC 121, the DPF 122, and the SCR 123 may be fluidly coupled by an exhaust gas conduit. The DOC 121 is structured to receive the exhaust gas from the engine 101 and to oxidize one or more exhaust gas constituents (e.g., hydrocarbons, carbon monoxide, etc.) in the exhaust gas.

[0024] The DPF 122 is arranged or positioned downstream of the DOC 121 and structured to remove particulates or particulate matter, such as soot, from exhaust gas flowing in theAtty. Dkt. No.: 106389-9772exhaust gas stream. The DPF 122 includes an inlet, where the exhaust gas is received, and an outlet, where the exhaust gas exits after having particulate matter substantially filtered from the exhaust gas. In some implementations, the DPF 122 or other components may be omitted and / or other components added (e.g., a second SCR system having an additional dosing unit or module, multiple DOCs, etc.). Additionally, although a particular arrangement is shown for the aftertreatment system 120 in FIG. 1, the arrangement of components within the aftertreatment system 120 may be different in other embodiments (e.g., the DPF 122 positioned downstream of the SCR 123 and ASC).

[0025] The aftertreatment system 120 may further include a reductant delivery system which may include a decomposition chamber (e.g., decomposition reactor, reactor pipe, decomposition tube, reactor tube, etc.) to convert a reductant into ammonia, shown as a dosing module or unit 124.

[0026] As shown, a plurality of sensors 125 are included in the aftertreatment system 120. The number, placement, and type of sensors included in the aftertreatment system 120 is shown for example purposes only. That is, in other configurations, the number, placement, and type of sensors may differ. The sensors 125 may be gas constituent sensors (e.g., NOx sensors, oxygen sensors, etc.), temperature sensors, particulate matter (PM) sensors, flow rate sensors (e.g., mass flow rate sensors, volumetric flow rate sensors, etc.), other exhaust gas emissions constituents sensors, pressure sensors, some combination thereof, and so on. The gas constituent sensors may include an oxygen sensor that is structured to acquire data indicative of the presence of oxygen in the exhaust gas. The data from the oxygen sensor may be used to estimate an AFR value. The flow rate sensors may include a mass air flow (MAF) sensor structured to acquire data indicative of a mass flow rate of the exhaust gas. The temperature sensors are structured to acquire data indicative of a temperature value at each location that the temperature sensor is located.

[0027] The sensors 125 may be located in or proximate the engine 101, after the engine 101 and before the aftertreatment system 120, after the aftertreatment system 120, in the aftertreatment system as shown (e.g., coupled to the DPF and / or DOC, coupled to the SCR, etc.), upstream of the engine 101, etc. It should be understood that the location of the sensors may vary. In one embodiment, there may be sensors 125 located both before and after theAtty. Dkt. No.: 106389-9772aftertreatment system 120. In one embodiment, at least one of the sensors is structured as exhaust gas constituent sensors (e.g., CO, NOx, PM, SOx, etc. sensors). In another embodiment, at least one of the sensors 125 is structured as non-exhaust gas constituent sensors that are used to estimate exhaust gas emissions (e.g., temperature, flowrate, pressure, etc.). Additional sensors may be also included with the system 100. The sensors may include engine-related sensors (e.g., torque sensors, speed sensors, pressure sensors, flowrate sensors, temperature sensors, etc.). For example, in some embodiments, at least one of the sensors 125 is structured as an oil temperature sensor that is used to detect and / or determine an engine oil temperature. The sensors may further include sensors associated with other components of the vehicle (e.g., speed sensor of a turbo charger, fuel quantity and injection rate sensor, fuel rail pressure sensor, etc.).

[0028] The sensors 125 may be real or virtual (i.e., a non-physical sensor that is structured as program logic that causes the controller to make various estimations or determinations). For example, an engine speed sensor may be a real or virtual sensor arranged to measure or otherwise acquire data, values, or information indicative of a speed of the engine 101 (typically expressed in revolutions-per-minute). The sensor is coupled to the engine (when structured as a real sensor), and is structured to send a signal indicative of the speed of the engine 101. When structured as a virtual sensor, at least one input may be used in an algorithm, model, lookup table, etc. to determine or estimate a parameter of the engine (e.g., power output, etc.). Any of the sensors 125 described herein may be real or virtual.

[0029] In some embodiments, the sensors 125 may include at least one of a temperature sensor or a pressure sensor. In other embodiments, the sensors 125 may include at least one of an oxygen sensor, a NOx sensor, and / or a particulate matter sensor. In still other embodiments, the sensors 125 may include other sensors, such as vehicle speed sensors (e.g., a tachometer, etc.) or other sensors configured to acquire data regarding the system 100.

[0030] The controller 150 includes at least one processing circuit having at least one processor and at least one memory device, and a communications interface. The controller 150 is configured to control operation of other components of the vehicle system 100 and / or the image recognition system 200.Atty. Dkt. No.: 106389-9772[00311 Now referring to FIG. 2, a schematic view of a block diagram of an image recognition system 200 used to assess the aftertreatment system 120 or component(s) thereof, among potentially other uses, is shown, according to an example embodiment. As shown, the image recognition system 200 includes an image capture device 140, a computing system 230, and a user device 240, among potentially other systems and / or components.

[0032] The image capture device 140 is configured or structured to acquire data regarding an object(s) of interest, such as of one or more components of the aftertreatment system 120, the engine, etc. In some embodiments, the image capture device 140 may be or include a borescope such that the image capture device 140 can acquire image data of the aftertreatment system 120 without disassembling the aftertreatment system 120 and / or removing one or more of the DOC 121, the DPF 122, and / or the SCR 123. For example, the image capture device 140 may be a borescope that can be inserted into the aftertreatment system 120 to acquire image data (e.g., pictures, videos, etc.) of a component or surface of a component of the aftertreatment system 120 (e.g., the DOC 121, the DPF 122, the SCR 123, etc.). In this manner, when the image capture device 140 is a borescope, the image capture device 140 can be inserted into the aftertreatment system 120 and the image recognition system 200 can acquire image data of the aftertreatment system 120 without disassembling the aftertreatment system 120.

[0033] In some embodiments, the image capture device 140 may be or include a digital image capture device capable of acquiring image data of the aftertreatment system 120. In this manner, the image capture device 140 can acquire image data (e.g., pictures, videos, etc.) of the aftertreatment system 120, and particularly, components thereof. In some embodiments, if the image capture device 140 is too large to be inserted into the aftertreatment system 120, the aftertreatment system 120 may be disassembled to acquire image data of the aftertreatment system 120. For example, if the image capture device 140 is too large to fit into the aftertreatment system 120, the aftertreatment system 120 may be disassembled or partially disassembled, and the image capture device 140 can acquire image data of the disassembled component or components.

[0034] In some other embodiments, the image capture device 140 may be or include any type of device configured to collect image data and send the image data to a different device, suchAtty. Dkt. No.: 106389-9772as a digital camera, a smartphone camera, etc. The image data may be or include video images which may be isolated to static images. That is, the image capture device 140 may collect, acquire, capture, receive, etc. video images and extract static images from the video images. Alternatively, the computing system 230 may analyze video images in addition to and / or in place of static images.

[0035] The image capture device 140 is coupled to the computing system 230, such that information may be exchanged between the image capture device 140 and the computing system 230, where the information may relate to one or more images regarding the aftertreatment system 120 and, particularly, components thereof. In some embodiments, the image capture device 140 is coupled with the computing system 230 via a wired connection. That is, a wire, such as a fiber optic cable, connects the image capture device 140 to the computing system 230 such that image data (e.g., pictures, videos) can be sent to the computing system 230 from the image capture device 140 along the wire. In some embodiments, the image capture device 140 may be coupled to the computing system 230 wirelessly, such that the image capture device 140 can send image data to the computing system 230 without requiring a physical, hardware connection between the image capture device 140 and the computing system 230. The image capture device 140 may be wirelessly coupled to the computing system 230 by, for example, Bluetooth®, such that the image capture device 140 can wirelessly send image data to the computing system 230 from the side of the shop opposite from the computing system 230.

[0036] In some embodiments, the computing system 230 may be embodied with or in the user device 240, such that when the image capture device 140 is coupled with the computing system 230, it is coupled with the user device 240. In this manner, the user device 240 may be directly coupled with the image capture device 140 such that the user device 240 directly interfaces with the image capture device 140 to send and receive information regarding operation of the image capture device 140 and / or image data acquired by the image capture device 140. For example, the user device 240 may be a laptop, a desktop, or other personal computing system that embodies the computing system 230. The user device 240 may be coupled to the image capture device 140 such that a command from the user device 240 causes the image capture device 140 to acquire image data. In some embodiments, the userAtty. Dkt. No.: 106389-9772device 240 may be or include a tablet, a laptop, a desktop computer, a mobile phone, or the like.[0O37| In some embodiments, the computing system 230 may be a remote computing system relative to the user device 240. In this manner, the computing system 230 may communicate with the image capture device 140 and receive image data from the image capture device 140. The computing system 230 may then analyze the received image data and thereafter communicate with the user device 240.

[0038] In some embodiments, the computing system 230 may be owned by, managed by, and / or otherwise associated with a provider entity. The provider entity may be an original equipment manufacturer (e.g., of the engine 101, parts or the system 100, etc.), a data analytics provider, a service provider (e.g., a repair or technician organization), a combination thereof, and so on. In some embodiments, the provider entity or organization may be remote from the system 100. In these embodiments, the computing system 230 may be a “remote” computing system. In some embodiments, a “remote” computing system may be a computing system that is located a predetermined distance away from the system 100. The computing system 230 includes at least one processing circuit 242 having at least one processor 246 and at least one memory device 248, one or more specialized circuits, shown as aftertreatment analysis circuit 250, and a communications interface 216.

[0039] The at least one processor 246 may be implemented as one or more single- or multichip processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), and / or suitable processors (e.g., other programmable logic devices, discrete hardware components, etc. to perform the functions described herein). A processor also may be implemented as a combination of computing devices, such as a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration. In some embodiments, the one or more processors 246 may be shared by multiple circuits (e.g., the aftertreatment analysis circuit 250 may include or otherwise share the same processor which, in some embodiments, may execute instructions stored, or otherwise accessed, via different areas or memory). Alternatively or additionally, the one or more processors 246 may be structured to perform or otherwise execute certain operationsAtty. Dkt. No.: 106389-9772independent of one or more co-processors. In other example embodiments, two or more processors may be coupled via a bus to enable independent, parallel, pipelined, or multithreaded instruction execution. All such variations are intended to fall within the scope of the present disclosure.

[0040] The at least one memory device 248 (e.g., memory, memory unit, storage device) may include one or more devices (e.g., RAM, ROM, Flash memory, hard disk storage) for storing data and / or computer code for completing or facilitating the various processes, layers, and modules described in the present disclosure. For example, the memory device 248 may include dynamic random-access memory (DRAM). The memory device 248 may be communicably connected to the processor 246 to provide computer code or instructions to the processor 246 for executing at least some of the processes described herein. Moreover, the memory device 248 may be or include tangible, non-transient volatile memory or nonvolatile memory. Accordingly, the memory device 248 may include database components, object code components, script components, or any other type of information structure for supporting the various activities and information structures described herein.

[0041] The communications interface 216 may include any combination of wired and / or wireless interfaces (e.g., jacks, antennas, transmitters, receiver, transceivers, wire terminals) for conducting data communications with various systems, devices, or networks structured to enable in-vehicle communications (e.g., between and among components of the vehicle) and out-of-vehicle communications (e.g., with a remote server). For example, and regarding out-of-vehicle / system communications, the communications interface 216 may include an Ethernet card and port for sending and receiving data via an Ethernet-based communications network and / or a Wi-Fi transceiver for communicating via a wireless communications network. The communications interface 216 may be structured to communicate via local area networks or wide area networks (e.g., the Internet) and may use a variety of communication protocols (e.g., IP, LON, Bluetooth, ZigBee, radio, cellular, near field communication). As shown in FIG. 2, the communications interface 216 is configured to enable communication with the image capture device 140 and the user device 240. For example, the communications interface 216 may enable the computing system 230 to transmit information, including one orAtty. Dkt. No.: 106389-9772more diagnoses and / or service actions, to a user device and / or receive a user input from the user device.[00421 In one configuration, the aftertreatment analysis circuit 250 is embodied as machine or computer-readable media storing instructions (e.g., programmable logic) that are executable by a processor, such as processor 246. The computer readable media instructions may include code, which may be written in any programming language including, but not limited to, Java or the like and any conventional procedural programming languages, such as the “C” programming language or similar programming languages. The computer readable program code may be executed on one processor or multiple remote processors. In the latter scenario, the remote processors may be connected to each other through any type of network (e.g., CAN bus, etc.).

[0043] In another configuration, the aftertreatment analysis circuit 250 is embodied as hardware units, such as one or more electronic control units. As such, the aftertreatment analysis circuit 250 may be embodied as one or more circuitry components including, but not limited to, processing circuitry, network interfaces, peripheral devices, input devices, output devices, sensors, etc. In some embodiments, the aftertreatment analysis circuit 250 may take the form of one or more analog circuits, electronic circuits (e.g., integrated circuits (IC), discrete circuits, system on a chip (SOCs) circuits, microcontrollers, etc.), telecommunication circuits, hybrid circuits, and any other type of “circuit.” In this regard, the aftertreatment analysis circuit 250 may include any type of component for accomplishing or facilitating achievement of the operations described herein. For example, a circuit as described herein may include one or more transistors, logic gates (e.g., NAND, AND, NOR, OR, XOR, NOT, XNOR, etc.), resistors, multiplexers, registers, capacitors, inductors, diodes, wiring, and so on). The aftertreatment analysis circuit 250 may include or be programmable hardware devices such as field programmable gate arrays, programmable array logic, programmable logic devices or the like. The aftertreatment analysis circuit 250 may include one or more memory devices for storing instructions that are executable by the processor(s) of the aftertreatment analysis circuit 250. The one or more memory devices and processor(s) may have the same definition as provided above with respect to the memory device 248 and processor 246. In some hardware unit configurations, the aftertreatment analysis circuit 250Atty. Dkt. No.: 106389-9772may be geographically dispersed throughout separate locations in the image recognition system 200. Alternatively, and as shown, the aftertreatment analysis circuit 250 may be embodied in or within a single unit / housing, which is shown as the computing system 230. In some embodiments, the aftertreatment analysis circuit 250 is configured to perform one or more methods described herein. The operations of the aftertreatment analysis circuit 250 are described in more detail herein with respect to FIGS. 3 and 4.[00441 The processing circuit 242 may be structured or configured to execute or implement the instructions, commands, and / or control processes described herein with respect to the aftertreatment analysis circuit 250. The depicted configuration represents the aftertreatment analysis circuit 250 being embodied as machine or computer-readable media storing instructions. However, as mentioned above, this illustration is not meant to be limiting as the present disclosure contemplates other embodiments where the aftertreatment analysis circuit 250 is configured as a hardware unit. All such combinations and variations are intended to fall within the scope of the present disclosure.

[0045] In some embodiments, the computing system 230 is embodied as a computing device, such as a desktop computer, a laptop computer, etc. In some other embodiments, the computing system 230 is embodied as a server, such as a cloud server, a privately owned server, a leased server (e.g. AWS, GCP, Azure, etc.), or another distributed computing system.

[0046] In some embodiments, the computing system 230 may be embodied in or proximate (e.g., within a predetermined distance of) the system 100. In some embodiments, the computing system 230 may be physically coupled to the system 100.

[0047] In some embodiments, the computing system 230 is configured to perform one or operations described herein. In some embodiments, the memory device 248 stores instruction that, when executed by the one or more processors 246, cause the one or more processors 246 to perform operations. In some embodiments, the memory device 248 may be or include non-transitory computer-readable media storing instructions that, when executed by one or more processors 246, cause the one or more processors 246 to perform operations.Atty. Dkt. No.: 106389-9772

[0048] In an example embodiment, the computing system 230 is structured to analyze image data regarding the aftertreatment system 120 to generate a diagnosis or a maintenance recommendation. The diagnosis and / or maintenance recommendation may be provided on a display, such as a display on the user device 240. In some embodiments, the diagnosis and / or the maintenance recommendation is provided to the user device 240 by message delivered via a messaging service (e.g., email, text message, etc.). In some embodiments, the message includes a hyperlink or a scannable code that directs a web browser of the user device 240 to a web-based version of the diagnosis and / or maintenance recommendation and / or a downloadable file (e.g., a PDF document, a .DOC(X) document, etc.) that includes the diagnosis and / or the maintenance recommendation. In other embodiments, the message includes a downloadable file that includes the diagnosis or the maintenance recommendation.

[0049] In still other embodiments, the message includes a downloadable file that includes the diagnosis or the maintenance recommendation and is provided via a user interface of a mobile application that is hosted by the user device 240. The mobile application may be a downloadable software application that is at least partially provided and supported by the computing system 230. In some embodiments, the generated user interface from the mobile application includes one or more predefined or pre-populated fields depicting one or more aspects of the diagnosis or the service action (information dynamically determined and / or other information associated and / or included with the diagnosis or the maintenance recommendati on) .

[0050] In some embodiments, the computing system 230 may generate two or more diagnoses and / or the maintenance recommendations. For example, the computing system 230 may generate a first maintenance recommendation (e.g., a near-term maintenance recommendation) and a second maintenance recommendation (e.g., an extended term maintenance recommendation). In another example, the computing system 230 may generate a first diagnosis (e.g., a diagnosis of a first DPF 122) and a second diagnosis (e.g., a diagnosis of a second DPF 122). That is, upon assessing the condition of the component of the aftertreatment system 120, the computing system 230 may generate a first maintenance recommendation providing instruction to remove the component of the aftertreatment system 120, and thereafter generate a second maintenance recommendation providing instruction toAtty. Dkt. No.: 106389-9772conduct maintenance on the component of the aftertreatment system 120 after a certain amount of time or distance. In another example, the computing system 230 may generate a first diagnosis indicating that the aftertreatment system 120 is not functioning properly and a second diagnosis indicating that the component of the aftertreatment system 120 should be removed and replaced.

[0051] In some embodiments, the user device 240 may be or include an external device configured to communicate with the computing system 230 regarding the image recognition system 200. For example, the user device 240 may be or include one or more of a smart watch, a smartphone, a laptop, a tablet, or the like. In this manner, the user device 240 may receive signals from the computing system 230 indicative of a state of the image recognition system 200, or the vehicle system 100. For example, the computing system 230 may send signals to the user device 240 indicating that the image capture device 140 is acquiring image data, signals indicating that the computing system 230 is assessing the aftertreatment system 120, and upon completion, the computing system 230 may send signals to the user device 240 causing a screen of the user device 240 to indicate the completion of the assessment and / or any relevant outcomes of the assessment. In some embodiments, the computing system 230 may send signals to the user device 240 causing the user device 240 to alert a user via an audible and / or visible notification.

[0052] In some embodiments, the user device 240 and the image capture device 140 are the same device. That is, the user device 240 and the image capture device are one device, such that the user device 240 can acquire image data. For example, the user device 240 and the image capture device 140 may be a smartphone with a camera such that the user device 240 and the image capture device 140 can acquire image data via the camera of the smartphone. In various other examples, the user device 240 and the image capture device may be a laptop with a camera, a smartwatch with a camera, a tablet with a camera, or the like.

[0053] In some embodiments, the user device 240 may be configured to receive one or more user inputs used to control operation of the image recognition system 200, and the components thereof. In some embodiments, the user device 240 may be a computing device associated with the computing system 230 (e.g., a terminal). The user device 240 may be coupled to the image capture device 140 such that the user device 240 can receive a userAtty. Dkt. No.: 106389-9772input and, in response to receiving the user input, cause the capture of image data by the image capture device 140, which is then sent from the image capture device 140 to the user device 240 and the computing system 230. The computing system 230 may execute one or more functions to assess the received image data, and upon completion, output results of the assessment to the user device 240.

[0054] In some embodiments, the user device 240 may be or include a graphical user interface configured to provide a plurality of graphical control elements to operate the image recognition system 200. In some embodiments, the user device 240 may include a touchscreen configured to receive touch inputs triggering the execution of various functions of the image recognition system 200. For example, upon positioning the image capture device 140 in a desired position, a graphical control element of the user device 240 configured to cause the image capture device 140 to capture one or more images of the field of view of the image capture device 140 may receive a touch input. The user device 240 may then provide a number of additional audio, visual, and / or audiovisual control elements configured to trigger various other functions of the image recognition system 200 to be executed thereafter. In some embodiments, the user device 240 may include a graphical user interface including a plurality of graphical control elements configured to receive an input to trigger various functions of the image recognition system 200. For example, having already received image data from the image capture device 140, the computing system 230 may cause the user device 240 to display a graphical control element associated with the execution of a function of the computing system 230. Upon the graphical control element being operated, the computing system 230 may execute the function identified by the graphical control element and output the result to the user device 240 with another graphical control element based on the output.10055] Now referring to FIG. 3, a flow diagram of a method 300 for assessing the aftertreatment system 120, or a component(s) thereof, is shown, according to an example embodiment. The computing system 230 and one or more components thereof is structured to perform the method 300. In particular, the aftertreatment analysis circuit 250 and one or more components thereof is structured to perform the method 300. It should be understood that the order of the method 300 is shown as example only. In some embodiments, one or more processes of the method 300 are optional. For example, a process 312 is optional and may beAtty. Dkt. No.: 106389-9772omitted from the method 300. In still other embodiments, one or more processes of the method 300 may be combined with one or more other depicted processes, and still further, additional processes may be added to the method 300 without departing from the spirit of the present disclosure. The method 300 may be performed periodically and / or dynamically responsive to receiving information, such as image data regarding a component of the aftertreatment system 120.]0056| At process 310, the computing system 230 receives image data of the aftertreatment system 120. In some embodiments, the computing system 230 may automatically cause the image capture device 140 to capture the image data. For example, the computing system 230 may determine a current field of view of the image capture device 140 is similar to a previously stored image of a surface of the aftertreatment system 120 and cause the image capture device 140 to capture one or more images of the current field of view of the image capture device 140. In other embodiments, the process 310 is conducted responsive to the receiving one or more instructions to acquire image data of the current field of view of the image capture device 140. These one or more instructions may be received by the computing system 230 from, for example, one or more of the image capture device 140 and / or the user device 240. In another example, the user device 240 receives a user input (e.g., a user operating a component of the user device 240) that causes the image capture device 140 to capture one or more images of the current field of view of the image capture device 140. In still another example, the image capture device 140 receives a user input (e.g., a user operating a component of the image capture device 140) that causes the image capture device 140 to capture one or more images of the current field of view of the image capture device 140.

[0057] In some embodiments, prior to the computing system 230 performing the process 310, the operator removes the component(s) being assessed and aligns the component with the field of view of the image capture device 140, and thereafter, the computing system 230 may perform the process 310. The user device 240 may provide instructions regarding the position of the component(s) relative to the image capture device 140 and instruct the operator where to position the component(s).Atty. Dkt. No.: 106389-9772[0058| In some other embodiments, prior to the computing system 230 performing the process 310, the user device 240 may provide the operator with instructions regarding operation of the borescope. The user device 240 may instruct the operator where to position the borescope prior to the process 310.

[0059] In some embodiments, the image capture device 140 captures one image of the current field of view of the image capture device 140. In some embodiments, the image capture device 140 captures multiple images of the current field of view of the image capture device 140. In some embodiments, the image capture device 140 may capture multiple images prior to the process 310 with different fields of view so as to capture image data with comprehensive coverage of the aftertreatment system 120 and / or one or more specific components being assessed, such as the DOC 121, the DPF 122, and / or the SCR 123.

[0060] The image data includes one or more images of one or more components of the aftertreatment system 120. For example, the image data may include an individual picture of the DPF 122. In another example, the image data may include a picture of the DOC 121. The components included in the image data include a number of features. In some embodiments, the features include a type of component included in the image data. For example, the features may indicate if the component is the DOC 121, the DPF 122, the SCR 123, or another component of the aftertreatment system 120. The features of the image data may correspond to one or more problems associated with the component. For example, when the image data includes an image of a soot accumulation on a surface of the DPF 122, the soot accumulation may correspond to one or more issues associated with the DPF 122. In some embodiments, the features include a color or an amount of deposits on the surface of the component of the aftertreatment system 120. For example, the one or more features of the image data may include an amount of the deposit on the component of the aftertreatment system 120 and a color of the deposit on the component of the aftertreatment system 120.

[0061] In some embodiments, the image data includes a first image of a surface of a component of the aftertreatment system 120. That is, the image data captured by the image capture device 140 and received by the computing system 230 may include a first image of the surface of the component of the aftertreatment system 120. According to an example embodiment, the component of the aftertreatment system 120 is at least one of the DPF 122,Atty. Dkt. No.: 106389-9772the DOC 121, or the SCR 123. That is, the component being assessed is at least one of the DPF 122, the DOC 121, or the SCR 123.[00621 At the process 310, the computing system 230 receives or acquires the image data. In some embodiments, the image data captured by the image capture device 140 is sent or transmitted to the computing system 230, such that the computing system 230 receives the image data to be used for assessing the aftertreatment system 120. In some embodiments, the image data is wirelessly received by the computing system 230 from the image capture device 140 via a wireless connection between the image capture device 140 and the computing system 230. In some embodiments, the image data is received by the computing system 230 from the image capture device 140 via a hardware (e.g., wired) connection. In some embodiments, responsive to the computing system 230 receiving the image data, the image recognition system 200 may proceed to process 312.[0063| In some embodiments, the process 310 includes the computing system 230 receiving a first image or a first series / plurality of images of a surface of a component of the aftertreatment system 120 from the image capture device 140. That is, the computing system 230 receives the first image or a first series / plurality of images of the surface of the component of the aftertreatment system 120 from the image capture device 140. In some embodiments, the process 312 is optional such that the image recognition system 200 may proceed from the process 310 to process 314 responsive to the computing system 230 receiving the image data.

[0064] At the process 312, the computing system 230 processes the image data received at the process 310. In some embodiments, the process 312 includes executing one or more operations to modify the image data to achieve at least one of enhancing the quality of the image or images contained in the image data compared to the received image data (e.g., by removing noise from the image or images contained in the image data, performing a color correction of the image or images contained in the image data, filtering the image or images contained in the image data to for example isolate an object of interest in the image(s) (e.g., the surface of the component under investigation), etc.), standardizing the format of the image or images contained in the image data, or resizing the image or images contained in the image data. That is, the process 312 may include a series of operations modifying the imageAtty. Dkt. No.: 106389-9772data, such as various operations to enhance the image data, removing noise from the image data, changing the format of the image data to standardize the image data, resizing the image data to be uniform and / or more easily analyzed by the computing system 230, conducting a filtering operation on the image data, and / or performing a color correction operation on the image data. For example, upon receiving the image data from the image capture device 140, the computing system 230 may apply a filter to the image data and modify the image data to increase the clarity and quality of an image such that the image data can be assessed more easily.10065] In some embodiments, resizing the image data includes resizing the image or images contained in the image data to a consistent resolution. In some embodiments, standardizing the format of the image data includes normalizing pixel values of the image or images of the image data to a standard scale (e.g., 0 to 1). In some embodiments, the process 312 includes removing unidentified features from the image or images of the image data and improving quality of the image or images of the image data. That is, the computing system 230 removes features that are not identified by the computing system 230.

[0066] In some embodiments, the process 312 includes at least one of removing unidentified features from the image data, removing noise from the image data, modifying the format of the image data, filtering the image data, and correcting a color of the image data. In some embodiments, responsive to processing the image data, the computing system 230 may provide the image data to a machine learning model and proceed to the process 314. In some embodiments, the process 312 is optional such that the image recognition system 200 may proceed directly from the process 310 to the process 314, providing the received image data to the machine learning model. The computing system 230 may employ a convolutional neural network or a type of image filter in order to remove and reduce noise in the image data. Modifying the format of the image data may be or include modifying a file type of the image data. In some embodiments, filtering and color correction of the image data may be or include modifying the visual appearance of the image data (e.g., the pixel color, size, clarity, etc.) in order to facilitate analysis of the image data by the computing system 230.

[0067] At the process 314, the computing system 230 extracts one or more features from the image data. In some embodiments, the process 314 is conducted by the machine learningAtty. Dkt. No.: 106389-9772model. In some embodiments, the feature extraction includes identifying, isolating, and characterizing one or more features captured in the image data. For example, the model may be configured to analyze the image data and extract features in the image data. That is, the feature extraction includes identifying one or more relevant features of the component of the aftertreatment system 120 (e.g., the DOC 121, the DPF 122, the SCR 123) captured in the image data and generating extracted images.]0068[ In some embodiments, the process 314 includes extracting a first feature from the image data. That is, the model may extract a first feature from the first image. For example, the model may identify the first feature in the first image, and isolate and characterize the first feature.

[0069] In some embodiments, the process 314 includes identifying, isolating and characterizing the deposit morphology and deposition patterns and distribution of particulates forming a special color and / or pattern on a surface of the aftertreatment system 120. That is, the computing system 230 uses a model that uses, as inputs, the shape and structure and the appearance caused by the deposit that has accumulated on the surface of the aftertreatment system 120 to identify, isolate, and characterize the surface deposits.

[0070] In some embodiments, the model utilized by the computing system 230 uses advanced image processing methods such as edge detection, texture analysis, and segmentation to execute the process 314 and extract features from the image data such as deposit morphology and deposition patterns and distribution of particulates forming special color and / or patterns. For example, while executing the process 314, the model may determine that the image data contains relevant deposit morphology and / or deposition pattern and distribution of particulates on the surface of the aftertreatment system 120 (e.g., the DOC 121, the DPF 122, the SCR 123) and proceed to isolate the portion of the image data containing the relevant features (i.e., the deposit morphology and / or the deposition pattern and distribution of particles), thereby creating an extracted image of the isolated features, and characterize the isolated features and the extracted images. According to an example embodiment, the model is a machine learning model configured to learn parameters from data to make the predictions, decisions, and determinations described herein.Atty. Dkt. No.: 106389-9772[00711 In some embodiments, the process 314 includes parsing a set of features from the image data and classifying each feature of the set of features. For example, the model may analyze a set of features, including a first feature, from the image data and classify each feature of the set of features.

[0072] In some embodiments, during the execution of the process 314, the model creates extracted images based on the extracted features. That is, the model generates one or more extracted images of the extracted features. In some embodiments, the extracted images include one or more images of at least one of the deposit morphology, the deposition patterns, or the distribution of particulates on the surface of the aftertreatment system 120. In some embodiments, responsive to extracting features from the image data and generating extracted images, the image recognition system 200 may proceed the process 316. In some embodiments, the images may be saved such that the images are retrievably accessible by the computing system 230 and / or the controller 150.

[0073] At the process 316, the computing system 230 analyzes the extracted images including, for example, the features extracted from the images. In some embodiments, the process 316 is executed by the machine learning model of the computing system 230. In some embodiments, the machine learning model is a file containing an algorithm that has been trained to recognize patterns in image data. The model can use the patterns to make predictions about new data. That is, the machine learning model includes a trained algorithm used to analyze and assess the image data. This algorithm and training process is further discussed below with reference to FIG. 4. In some embodiments, the process 316 includes the model using the extracted image or images to recognize and classify different types of deposits based on the deposit distribution pattern or color. The image recognition system 200, and more particularly, the computing system 230, receives the extracted images and compares features contained in the extracted images to a set of historical features stored in the trained machine learning model to identify patterns and relationships contained in the extracted image or images.[0074J In some embodiments, the process 316 includes identifying a second feature that corresponds to the first feature based on comparing the first feature to the set of historical features, the set of historical features including the second feature. That is, the extractedAtty. Dkt. No.: 106389-9772images are compared to the set of historical features to identify a second feature that corresponds to the first feature. For example, the model may generate extracted images as described above and compare the extracted images to the set of historical data to determine the condition of the surface of the aftertreatment system 120.

[0075] “Correspondence” or “corresponds” is used herein to describe one feature aligning, relating, matching, and / or substantial matching to another feature, such as between the second feature and first feature of the image data. For example, corresponding features may be determined based on analysis of the pixels in in the image data. As a specific example, a first feature corresponding to a second feature may be based on a pixel or group of pixels in a first image of the image data indicating the same or similar color or other image data characteristic as a pixel or group of pixels in a second image. The corresponding or aligning pixels or group of pixels may be associated with first and second features, respectively, that are then determined by the computing system to correspond to each other (e.g., match, substantially match, etc.). The pixel or grouping of pixels may indicate locations of certain conditions, such as soot or other particulate matter build-up on the filter (e.g., DPF). Further, based on where the features are identified, the computing system may determine potential failure modes based on historical data regarding soot build-up (based on a pixel analysis) in those location(s) of the filter element or object. That way, the computing system 230 may quickly identify potential undesirable conditions relatively quickly.

[0076] The “condition,” as used to describe a component s), such as a surface of the aftertreatment system 120 component, refers to a characteristic of the component(s). The condition may affect the functionality of the component(s). Thus, the condition may include an indication of or description of a deposit on a filter, such as a surface area value of the deposit on the filter (e.g., how much area is covered by the deposit), a thickness of a deposit in various areas on the filter, a type of material that forms the deposit, an indication of an age of the component, an indication of the presence of other abnormalities in the component(s) (e.g., cracks, holes, etc.), etc. The condition may be used to determine whether maintenance or examination may be recommended.

[0077] In some embodiments, identifying the patterns and relationships contained in the extracted image or images includes identifying a quantitative assessment of the depositAtty. Dkt. No.: 106389-9772coverage on the surface of the aftertreatment system 120, a type of deposit, a condition of the surface of the aftertreatment system 120, a pattern-based failure mechanism, and detailed failure modes. That is, the machine learning model receives the extracted image or images, compares the extracted image or images to the set of historical features in the model, and outputs results of the comparison, thereby diagnosing the surface of the aftertreatment system 120.[0078| In some embodiments, the quantitative assessment of the deposit coverage includes an assessment of a degree of coverage of the surface of the aftertreatment system 120 by the deposit. That is, the computing system 230, using the model, may determine an amount (e.g., quantity, surface area, volume, or density) of the contaminant on of the surface of the aftertreatment system 120 component. For example, based on the correspondence between the first feature and the second feature, the computing system 230, using the model, may determine an amount of coverage of the surface of the aftertreatment system 120 component by a contaminant and quantify the amount of coverage as the degree of coverage, which may be expressed as a quantity (e.g., mass, weight, molar amount), surface area, volume, or density.[0079| In some embodiments, identifying the type of deposit includes identifying a contaminant on the surface of the aftertreatment system 120. That is, the model may identify the contaminant on the surface of the aftertreatment system 120 based on the correspondence between the first feature and the second feature. For example, the model may determine that the characteristics of the extracted image resemble a soot deposit, and determine the contaminant to be soot.[0()80| In some embodiments, identifying the condition of the surface of the aftertreatment system 120 includes identifying the age, degree of wear, usability, and / or remaining life span of the component of the aftertreatment system 120 containing the surface being assessed. That is, the computing system 230, particularly the image recognition system 200, identifies the current condition of the surface of the aftertreatment system 120 and generates values regarding an age of the component and / or a degree of wear, and based on the degree of wear, an indication regarding the usability of the component and a value of the remaining life span of the component. For example, based on the correspondence between the first feature andAtty. Dkt. No.: 106389-9772the second feature, the model may determine information the component of the aftertreatment system 120 such as the age, degree of wear, usability, and / or remaining life span.[00811 In some embodiments, the failure mechanism is an indication of one or more causes of the failure of the component of the aftertreatment system 120. That is, the computing system 230, particularly the image recognition system 200, identifies one or more causes of the failure of the component of the aftertreatment system 120 based on the correspondence between the first feature and the second feature. In some embodiments, the failure mode is the way in which the component of the aftertreatment system 120 fails. That is, the image recognition system 200 identifies one or more ways in which the component of the aftertreatment system 120 has failed, is failing, or is expected to fail based on the correspondence between the first feature and the second feature.

[0082] At the process 318, the computing system 230 assesses the condition of the surface of the component of the aftertreatment system 120 (e.g., the surface of the DOC 121, the DPF 122, and / or the SCR 123). In some embodiments, the process 318 is executed by the computing system 230. In some embodiments, assessing the condition of the surface of the aftertreatment system 120 includes generating, based on the determination from the machine learning model, a report detailing the condition on the filter surface. That is, the computing system 230 uses the output from the machine learning model to generate a report of the condition of the surface of the aftertreatment system 120. In some embodiments, the report of the condition of the surface of the aftertreatment system 120 includes the quantitative assessment of the deposit coverage, the identification of the prominent deposit type, the identified pattern-based failure mechanism and the detected failure mode. Upon generating the report, the computing system 230 may complete the assessment of the condition of the surface of the aftertreatment system 120, wherein the surface is at least one of the DOC 121, the DPF 122, and the SCR 123.

[0083] In some embodiments, the process 318 includes generating, based on at least one of the determinations of the model or the comprehensive report, a maintenance recommendation. That is, the computing system 230 may generate a maintenance recommendation based on features of the image data. For example, responsive to receiving the determination from the model indicating that the component is damaged and needsAtty. Dkt. No.: 106389-9772maintenance, the computing system 230 may generate a maintenance recommendation detailing operations to be done to the component to resolve the damage.

[0084] In some embodiments, the computing system 230 generates the maintenance recommendation based on at least one of the output of the machine learning analysis model and / or the comprehensive report of the process 318. That is, the maintenance recommendation is based on at least one of the results of analyzing the extracted image or images and / or the report generated during the execution of the process 318. In some embodiments, the maintenance recommendation includes various relevant data regarding the condition of the component of the aftertreatment system 120 being assessed. In some embodiments, the relevant data included in the maintenance recommendation may be or include a number of actionable recommendations, issue descriptions, probable failure modes, quantitative efficiency, usable life left, and other parameters regarding the assessment of the component of the aftertreatment system 120. For example, the maintenance recommendation may include one or more actionable recommendations for maintenance of the aftertreatment system 120 and the component of the aftertreatment system 120 being assessed. In some embodiments, there may be a lookup table correlating the condition to a maintenance recommendation based on the condition. The maintenance recommendation may then be retrieved and included in the report.

[0085] In some embodiments, the actionable recommendations of the maintenance recommendation can be or include one or more of a “no maintenance required” recommendation, indicating that the component of the aftertreatment system 120 is operational in a manner that it satisfies its respective performance requirements and does not need to be removed or replaced, a “replace” recommendation, indicating that the component of the aftertreatment system 120 needs to be removed and replaced in order for the aftertreatment system 120 to meet performance requirements or to ensure proper operation of the vehicle system 100, or a “send to lab” recommendation indicating that the component of the aftertreatment system 120 needs to be removed and sent to a testing lab for further analysis, and also needs to be replaced.

[0086] In some embodiments, the maintenance recommendation may be or include one or more issue descriptions relating to one or more issues of the surface of the aftertreatmentAtty. Dkt. No.: 106389-9772system 120. That is, the maintenance recommendation may include a detailed description of each identified feature of the image data. In some embodiments, the issue descriptions are based on one or more issues indicated in the report generated in the process 318. For example, the image recognition system 200 may determine that the component of the aftertreatment system 120 has too much deposit accumulated on its surface and document that issue in the report generated during the process 318.]0087| In some embodiments, the maintenance recommendation includes one or more probable failure modes regarding the way the component of the aftertreatment system 120 fails or may fail. The one or more probable failure modes of the maintenance recommendation may be or include a plain English recitation of a failure mode identified and indicated in the process 316 and the process 318 respectively. That is, the maintenance recommendation may be or include an indication of the way in which the component of the aftertreatment system 120 has failed, is failing, or will fail, based on the process 316 and the comprehensive report of the process 318.

[0088] In some embodiments, the maintenance recommendation includes an indication of a quantitative efficiency of the component of the aftertreatment system 120. The quantitative efficiency may be or include an indication of the efficacy of the component of the aftertreatment system 120 in its current state of operation relative to when the component of the aftertreatment system 120 was new. That is, the maintenance recommendation may include an indication of the current performance of the component of the aftertreatment system 120 compared to the performance of the component of the aftertreatment system 120 when the component of the aftertreatment system 120 is healthy. In some embodiments, the quantitative efficiency may be a numerical value indication how effective the component of the aftertreatment system 120 is at performing its respective function in the aftertreatment system 120. The efficiency of the component may be received from the controller 150 for example. The controller 150 may determine an efficiency of the component based on the current capabilities (e.g., a current NOx conversion efficiency) of the component (e.g., SCR) compared to the initial capabilities (e.g., an initial or starting NOx conversion efficiency) of the component (e.g., SCR).Atty. Dkt. No.: 106389-9772[0089| In some embodiments, the maintenance recommendation includes an indication of the usable life left of the component of the aftertreatment system 120. That is, the maintenance recommendation may be or include an indication of how much longer the component of the aftertreatment system 120 can continue to be used before the component of the aftertreatment system 120 fails. In some embodiments, the usable life left of the component of the aftertreatment system 120 is based on the life span indicated during the process 316. In some embodiments, the usable life left of the component of the aftertreatment system 120 may be or include a distance or time value (e.g., mileage value indicating how many more miles the component can be used for before a potential failure occurs). In some embodiments, the usable life left of the component of the aftertreatment system 120 may be or include a time value indicating how much longer the component can remain in the aftertreatment system 120 before the component likely is no longer capable of satisfying one or more performance criteria.

[0090] In some embodiments, the maintenance recommendation may be or include a prompt to schedule preventative maintenance activities and / or replacements based on the conditions. That is, the maintenance recommendations, based on the surface condition of the component of the aftertreatment system 120 may include a prompt to schedule future preventative maintenance to be performed on the component of the aftertreatment system 120. The prompt may be or include text, provided via a graphical user interface of the user device 240 in, for example, the report.[0091 J At the process 320, the computing system 230, particularly the image recognition system 200, provides the maintenance recommendation. In some embodiments, the process 320 is executed by the computing system 230 in combination with the user device 240. That is, the computing system 230 and the user device 240 work in combination to provide the maintenance recommendation. For example, the computing system 230 may output the maintenance recommendation to a graphical interface of the user device 240. In some embodiments, the process 320 includes generating a notification for display on a user interface, the notification indicating the condition of the component of the aftertreatment system 120. That is, the process 320 may include generating a notification based on orAtty. Dkt. No.: 106389-9772including the maintenance recommendation and regarding the condition of the component of the aftertreatment system 120.[00921 In some embodiments, the maintenance recommendation may be provided to the user device 240 by the computing system 230. That is, the user device 240 may receive the maintenance recommendation from the computing system 230, and thereafter provide the maintenance recommendation via a graphical user interface of the user device 240.

[0093] In some embodiments, the user device 240 provides a graphical representation of at least one of the maintenance recommendation or the comprehensive report. The graphical representation may be displayed on a user interface of the user device 240 to communicate the diagnosis of the component and / or how to resolve the diagnosis via the maintenance recommendation. In some embodiments, the graphical representation may include a written description of the maintenance recommendation and / or the comprehensive report. In other embodiments, the graphical representation may include a visual step-by-step description detailing how to perform the maintenance recommendation. For example, the graphical representation may include a flowchart, work instructions, a service process diagram, an instructional video, or the like.

[0094] As discussed above, the method 300 can eliminate or reduce the need for manual inspection and analysis of the surface of the component of the aftertreatment system 120 and conduct the assessment of the component of the aftertreatment system 120 faster than the manual inspection and analysis. Advantageously, these capabilities of the method 300 can reduce the cost of and increase the efficiency of the assessment, by eliminating manual inspection and conducting the assessment more quickly than the manual inspection and assessment.[00951 Now referring to FIG. 4, a flow diagram of a method 400 for training the machine learning model is shown, according to an example embodiment. In particular, the computing system 230 and / or one or more components thereof is structured to perform the method 400. It should be understood that the order of the method 400 is shown as example only. In some embodiments, one or more of the processes of the method 400 are optional. For example, a process 412 is optional and may be omitted from the method 400. In still other embodiments,Atty. Dkt. No.: 106389-9772one or more of the processes of the method 400 may be combined with one or more other depicted processes, and still further, additional processes may be added to the method 400 without departing from the spirit and scope of the present disclosure.

[0096] In some embodiments, the method 400 is conducted to train the machine learning model of the computing system 230. For example, when the computing system 230 executes the method 400, the computing system 230 trains the machine learning model will, which, in turn, causes the machine learning model to assess components of the aftertreatment system 120 more accurately. In some embodiments, the method 400 may be executed multiple times to increase the accuracy and reliability of the machine learning model to a desired level.

[0097] At process 410, the computing system 230 receives image data. In some embodiments, the computing system 230 receives the image data via a user input. In some embodiments, the image data includes various types of surface images, with labeled descriptions corresponding to each image. For example, the image data received at the process 410 may be or include images similar to the images shown in FIGS. 5-8, each of the images labeled with a detailed description of the image. In some embodiments, the image data includes images with different types of contaminants, damages, and anomalies relevant to surfaces of the aftertreatment system 120. The different types of contaminants, damages, and anomalies can correspond to detailed descriptions of failure mode and actions to be taken contained in the description of the image.

[0098] As shown in FIGS. 5-8, the image data received at the process 410 includes images of healthy (i.e., fully operational) surfaces of the aftertreatment system 120 and images of unhealthy (i.e., not fully operational) surfaces of the aftertreatment system 120, each labeled in a form that the machine learning model can train on and learn from. In some embodiments, the description of each of the images of the image data includes a failure mode detailing a way that the component fails, a root cause for the failure indicating a deposit which has accumulated on the surface of the component, a unit of time used to measure the probability of the component failing dangerously or the rate at which a component fails (i.e., a failure hour), a remedy to reverse the effects of the deposit on the surface, and a strategy for avoidance for a future failure or defect of the surface captured in the image.Atty. Dkt. No.: 106389-9772[0099| In some embodiments, the failure hour of the component may be or include an indication of the remaining usability of the component, providing an indication of how much longer the component can be used before it fails. In some embodiments, the failure hour may be or include a prediction of the probability that the component will fail. That is, the failure hour may be or include an indication of how much longer the component will be operational and / or an indication of the probability of the component failing. For example, the failure hour may be an indication that the component will be usable for a certain amount of time and / or distance before it is likely to fail, with an indication of how likely the failure is.10100] In some embodiments, the remedy may be or include actionable steps to be taken to improve the condition of the surface of the component of the aftertreatment system 120. For example, the remedy may provide instruction to clean the surface to restore the function of the component. In another example, the remedy may provide instruction to remove and replace the component. That is, the remedy may be or include instructions to improve the condition of the component of the aftertreatment system 120 that, when executed, improve or protect the function of the aftertreatment system 120.

[0101] In some embodiments, the strategy for avoidance may be or include instructions to schedule future maintenance after a predetermined amount of time or use of the vehicle system 100. For example, the strategy for avoidance may provide instruction to conduct maintenance on the aftertreatment system 120 every 5,000 miles of use of the vehicle system 100. That is, the strategy for avoidance may include instructions to schedule maintenance on the aftertreatment system 120 at a certain time based on the condition of the surface of the component of the aftertreatment system 120.[01021 In some embodiments, the description of each of the images includes any number of the failure mode, the root cause, the failure hour, the remedy, and the strategy for avoidance. That is, the description of each of the images of the image data may include any number of the failure mode, the root cause, the failure hour, the remedy, and the strategy for avoidance. For example, the description of each image may include any one of the failure modes, the root cause, the failure hour, the remedy, and the strategy for avoidance. In another example, the description of each image may include any two of the failure mode, the root cause, the failure hour, the remedy, and the strategy for avoidance. In various other examples, theAtty. Dkt. No.: 106389-9772description of each image may include any number of the failure mode, the root cause, the failure hour, the remedy, and the strategy for avoidance.[01O3| In some embodiments, the description of each of the images includes corresponding labels indicating the condition or maintenance requirement of the surface contained in the image. For example, the description may include condition identifiers and maintenance instructions such as “clean,” “replace,” “damaged,” “clogged,” or the like. In some embodiments, responsive to receiving the image data and determining that each image includes a description, the computing system 230 may proceed to process 412. In some embodiments, the procession from the process 410 to the process 412 is optional, and the computing system 230 may instead proceed from the process 410 to process 414.

[0104] At the process 412, the computing system 230 processes the received image data. In some embodiments, the processing of the image data includes improving each of the images to be more easily and accurately assessed by the model. In some embodiments, processing the image data includes resizing images, normalizing pixel values, augmenting the image data, and removing outliers. Resizing the image data may be or include various operations to ensure each of the images are a consistent resolution, thereby facilitating training of the machine learning model by ensuring each of the images has the same resolution. Normalizing pixel values of each of the images may be or include setting the pixel values to a standard scale (e.g., 0 to 1) to improve convergence training. Augmenting the image data may include employing techniques such as rotation, flipping, and adding noise to increase diversity in the image data and improve model generalization. The process 412 may include removing any outliers or artificial, non-related failure images from the image data to improve consistency. Responsive to processing the image data, the computing system 230 may proceed to process 414.

[0105] At the process 414, the computing system 230 selects a machine learning model architecture. In some embodiments, selecting the model architecture includes choosing a suitable deep learning architecture for image recognition tasks. In some embodiments, the model architecture will be selected via, for example, a user input received by the image recognition system 200 at the user device 240. For example, the computing system 230 may receive a user input indicating a Convolutional Neural Network (CNN). In someAtty. Dkt. No.: 106389-9772embodiments, the computing system 230 may contain a number of various pre-trained models or custom architectures tailored to the specific requirements of surface analysis, and automatically select one of the various models or architectures. For example, the computing system 230 may receive image data, and automatically select a trained machine learning model stored in the memory device 248. In some embodiments, upon completion of the process 414, the computing system 230 may proceed to process 416.]0106| At the process 416, the computing system 230 operates the machine learning model to execute a transfer learning technique. In some embodiments, the process 416 includes the model selected in the process 414 being initialized with pre-trained weights on large-scale image data to leverage learned features and accelerate convergence. That is, the computing system 230 may start the selected model architecture and input a large set of image data to rapidly train and improve the model. In some embodiments, the process 416 is an iterative process. For example, the process 416 may be executed multiple times sequentially, with the operator setting goals for the performance of the model, inputting image data, testing, reviewing and tuning parameters of the model to more accurately recognize surface-specific features and conditions, and repeating the process until the model reaches the performance goals consistently, or at a predefined rate. In some embodiments, responsive to the process 416 being completed, the computing system 230 may proceed to process 418.|0107] At the process 418, the computing system 230 trains the model. In some embodiments, the process 418 includes the model receiving the image data and splitting the image data into various data sets to evaluate the model performance accurately. For example, the data sets may include a training set used to train the model, a validation set used to evaluate the performance of the model and tune the parameters of the model, and a test set used to evaluate the performance of the model after the model has been tuned by the validation set.

[0108] The training set may be presented to the model with both the images and the associated descriptions to train the model to recognize patterns based on the description and visual indicators in the image data. Each image of the validation set may be presented to the model without the description of the image, with the description provided after the model has analyzed its associated image, with the model comparing the description to its analysis andAtty. Dkt. No.: 106389-9772using the results to develop a pattern, learn, and evaluate the performance of the model. The test set may be presented to the model without the description, with the description being provided after completing analysis of the entire set to analyze the performance of the model.

[0109] For example, the image data may contain 1,300 images. 700 of the images may be separated into the training set, with 300 images being separated into the validation set, and the remaining 300 images being separated into the test set. The model may first receive and analyze the 700 images of the training set, identifying patterns between visual indicators of each image and the corresponding description. Upon completion of the analysis of the training set, the model may receive the 300 images of the validation set, analyzing each image based on the visual indicators in the image data, receiving the description corresponding to each image and comparing the analysis of the model to the description, analyzing the accuracy of the analysis and tuning the model before proceeding to the next image of the validation set. Responsive to completing the analysis of the validation set, the model may receive the 300 images contained in the test set. The model may then analyze each image of the test set. Responsive to analyzing each image of the test set, the model may receive the descriptions associated with each of the images of the test set and compare the analysis of each image to the description associated with each image to assess the performance of the model.1'011.0] In some embodiments, the execution of the process 418 as described above may include the use of a suitable optimization algorithm and a loss function appropriate for multiclass classification tasks. Throughout the execution of the process 418, the computing system 230 may cause the model to monitor training progress by tracking accuracy, precision, recall, and generating an Fl score of the model based on the validation set. The Fl score may be used by the computing system 230 to prevent overfitting by the model and adjust hyperparameters of the model to improve the performance of the model. In some embodiments, the process 418 includes hyperparameter tuning for the model. That is, the computing system 230 may experiment with different hyperparameters of the model such as learning rate, batch size, dropout rate, and network depth to improve model performance. The hyperparameter tuning may include utilizing techniques such as grid search or random search to explore a hyperparameter space and identify optimal configurations. In someAtty. Dkt. No.: 106389-9772embodiments, responsive to training the model to desired performance levels and / or hyperparameter tuning levels, the computing system 230 may proceed to process 420.[01111 At the process 420, the computing system 230 evaluates the accuracy of the model. In some embodiments, evaluating the model accuracy includes using the 300 images of the test set and analyzing the visual indicators found in the images to develop an assessment of the condition of the surface of the aftertreatment system 120. The model may analyze each image of the test set and thereafter receive all the associated descriptions for the images of the test set. Upon receiving the associated descriptions, the model may assess the performance and robustness of the model by comparing the associated descriptions to the assessments generated by the model. In some embodiments, the process 420 includes calculating metrics such as accuracy, precision, recall, and Fl score, which are used to quantify the ability of the model to correctly classify surface conditions of the aftertreatment system 120. That is, the model may develop various metrics to assess the performance of the model. In some embodiments, the process 420 may include conducting error analysis regarding the performance of the model and the accuracy of the assessment to identify common misclassifications and areas for improvement in the model. Responsive to evaluating the accuracy of the model, the computing system 230 may proceed to process 422.

[0112] At the process 422, the computing system 230 tunes the model according to various parameters. In some embodiments, the process 422 includes modifying parameters such as model architecture, hyperparameters, and training strategies associated with the model based on insights gained in the process 420. That is, the computing system 230 may use the evaluation of the model, the various metrics associated with the evaluation of the model, and the error analysis of the process 420 to tune parameters of the model to improve the performance of the model relative to the prior parameters. In some embodiments, the process 422 includes using techniques such as model pruning, quantization, or compression to reduce model size and computational complexity to allow for deployment on resource-constrained devices.

[0113] In some embodiments, the process 422 includes optimizing the model. In some embodiments, the process 422 includes returning to execute the process 414 to select an alternative model architecture based on the outcome of the process 420. That is, theAtty. Dkt. No.: 106389-9772computing system 230 may select an alternative model architecture to improve the performance of the model relative to the architecture previously selected at the process 414. In some embodiments, the process 422 includes returning to execute the process 418 to tune hyperparameters of the model. That is, the computing system 230 may modify hyperparameters of the model based on the outcome of the process 420 to improve the performance of the model relative to the model using the previous hyperparameters. In some embodiments, responsive to tuning the model and preparing the model for deployment, the computing system 230 may continue to process 424.10114] At the process 424, the model is fully trained and is deployed to a target destination. In some embodiments, the process 424 includes integrating the model into the computing system 230 and / or another component of the image recognition system 200. For example, the method may be deployed to the computing system 230. This may include the model being accessible via the user device 240 such that a user can send and receive signals and information to the model via the user device 240 and the computing system 230. During the process 424, the model is confirmed to be capable with an image processing module and a maintenance recommendation engine of the image recognition system 200. In some embodiments, the model is deployed in production environments. For example, the model may be deployed in a production environment either on-premises or in the cloud, to enable real-time condition monitoring and maintenance for the aftertreatment system 120 and the components thereof such as the DOC 121, the DPF 122, and the SCR 123.

[0115] In some embodiments, the method 400 can allow for training and improvement of the machine learning model of the image recognition system 200 without requiring manual training. That is, the computing system 230 can eliminate the need for the operator to modify parameters of, evaluate accuracy of, and otherwise train the model. Advantageously, the elimination of operator training may allow the model to learn and train more quickly and become accurate and reliable more quickly than it would have had an operator trained it.

[0116] Now referring to FIGS. 5-8, various views of a surface of the DPF 122 are shown, according to an example embodiment. In some embodiments, FIGS. 5-8 may be included in a set of image data received by the image recognition system 200 in the method 300 and / or the method 400. Now referring to FIG. 5, an image 500 of a view of a clean surface of the DPFAtty. Dkt. No.: 106389-9772122 is shown, according to an exemplary embodiment. As shown in the image 500, the DPF 122 is clean, with open pores and no deposit accumulation or other defects. In some embodiments, a description associated with the image 500 may include an indication of the condition of the surface of the DPF 122 in the image 500. For example, the description may include plain English such as “Clean and open pores - fresh DPF surface,” and be received by the image recognition system 200 to characterize the image 500.[0117{ In some embodiments, the image recognition system 200 may receive the image 500 as shown in FIG. 5, with no accompanying description. For example, during a training operation as described above with reference to the process 418 and the process 420, the image recognition system 200 may receive the image 500 with no accompanying description to train or evaluate the model. In some embodiments, the image recognition system 200 may receive an image similar to the image 500 by, for example the image capture device 140 while conducting the method 300, wherein the image 500 is not accompanied by a description. For example, the image recognition system 200 may receive the image 500 and execute the method 300 to analyze the image 500 and diagnose the surface of the DPF 122. In some embodiments, the image 500 may be accompanied by a corresponding description. In some embodiments, the image recognition system 200 may receive the image 500 and the accompanying description. For example, the image recognition system 200 may receive the image 500 and the accompanying description and execute the method 400 to train the model.

[0118] Now referring to FIG. 6, an image 600 of a soot distribution on a surface of the DPF 122 is shown, according to an example embodiment. As shown in FIG. 6, the DPF 122 has a soot distribution pattern with densely packed soot. In some embodiments, a description associated with the image 600 may include an indication of the condition of the surface of the DPF 122 in the image 600. For example, the description may include plain English such as “Soot distribution pattern with densely packed soot,” and be received by the image recognition system 200 to characterize the image 600. The description of the image 600 may be or include various indications regarding the amount of deposit accumulation on the surface of the DPF 122. For example, the description of the deposit may include quantitative descriptions regarding the deposit such as a mass of the deposit, an amount of surface area covered by the deposit on the surface of the DPF 122, a volume of the deposit on the surfaceAtty. Dkt. No.: 106389-9772of the DPF 122, or the like. The description of the image 600 may be or include various indications regarding the color of the deposit on the surface of the DPF 122. For example, the description of the deposit may include qualitative descriptions indicating the color of the deposit on the surface of the DPF 122, such as a hex code of the color of the deposit on the surface of the DPF 122, an RGB value of the color of the deposit on the surface of the DPF 122, a pantone color value of the DPF 122, or the like. The description of the image 600 may include qualitative descriptions indicating the pattern of each of the deposits on the surface of the DPF 122. For example, the description may include a location of the deposit on the surface of the DPF 122, a composition of the deposit throughout its volume, an indication of the concentration of the deposit, one or more qualitative explanations of a pattern of the deposit on the surface of the DPF 122, or the like.[0119| Now referring to FIG. 7, an image 700 of a foreign contaminant accumulation on a surface of the DPF 122 is shown, according to an example embodiment. As shown in the image 700, the surface of the DPF 122 includes a foreign contamination. In some embodiments, a description associated with the image 700 may include an indication of the condition of the surface of the DPF 122 in the image 700. For example, the description may include plain English such as “Foreign contamination distinguished by a colored accumulation,” and be received by the image recognition system 200 to characterize the image 700. In the example shown, the foreign contamination is shown by a colored accumulation on the DPF 122, shown in FIG. 7 as the box pattern section, particularly the diamond-shaped cross hatching section, of the image 700 (located proximate the left-hand side of the image). The description of the image 700 may be or include various indications regarding the amount of deposit accumulation on the surface of the DPF 122. For example, the description of the deposit may include quantitative descriptions regarding the deposit such as a mass of the deposit, an amount of surface area covered by the deposit on the surface of the DPF 122, a volume of the deposit on the surface of the DPF 122, or the like. The description of the image 700 may be or include various indications regarding the color of the deposit on the surface of the DPF 122. For example, the description of the deposit may include qualitative descriptions indicating the color of the deposit on the surface of the DPF 122, such as a hex code of the color of the deposit on the surface of the DPF 122, an RGB value of the color of the deposit on the surface of the DPF 122, a pantone color value of theAtty. Dkt. No.: 106389-9772DPF 122, or the like. The description of the image 700 may include qualitative descriptions indicating the pattern of each of the deposits on the surface of the DPF 122. For example, the description may include a location of the deposit on the surface of the DPF 122, a composition of the deposit throughout its volume, an indication of the concentration of the deposit, one or more qualitative explanations of a pattern of the deposit on the surface of the DPF 122, or the like. For example, the dot pattern section of the image 700 may be or include soot accumulation on the surface of the DPF 122 (proximate the right-hand side of the image 700), while the diamond or box pattern section of the image 700 may be or include accumulation of a foreign contaminant on the surface of the DPF 122. Descriptions of the dot pattern section and the box pattern section of the image 700 may be provided in the description of the image 700.[0120| Referring now to FIG. 8, an image 800 of a locally concentrated soot accumulation on a surface of the DPF 122 is shown, according to an example embodiment. As shown in the image 800, the surface of the DPF 122 includes a concentrated soot accumulation. In some embodiments, a description associated with the image 800 may include an indication of the condition of the surface of the DPF 122 in the image 800. For example, the description may include plain English such as “Locally concentrated soot accumulation,” and be received by the image recognition system 200 to characterize the image 800. In some embodiments, the soot accumulation is shown by the concentrated deposit on the surface of the DPF 122. The description of the image 800 may be or include various indications regarding the amount of deposit accumulation on the surface of the DPF 122. For example, the description of the deposit may include quantitative descriptions regarding the deposit such as a mass of the deposit, an amount of surface area covered by the deposit on the surface of the DPF 122, a volume of the deposit on the surface of the DPF 122, or the like. The description of the image 800 may be or include various indications regarding the color of the deposit on the surface of the DPF 122. For example, the description of the deposit may include qualitative descriptions indicating the color of the deposit on the surface of the DPF 122, such as a hex code of the color of the deposit on the surface of the DPF 122, an RGB value of the color of the deposit on the surface of the DPF 122, a pantone color value of the DPF 122, or the like. The description of the image 800 may include qualitative descriptions indicating the pattern of each of the deposits on the surface of the DPF 122. For example, the description may includeAtty. Dkt. No.: 106389-9772a location of the deposit on the surface of the DPF 122, a composition of the deposit throughout its volume, an indication of the concentration of the deposit, one or more qualitative explanations of a pattern of the deposit on the surface of the DPF 122, or the like.

[0121] In some embodiments, images similar to the images 500-800 may be received by the image recognition system 200 and used to develop pattern-identifying in the image recognition system 200. That is the image recognition system 200 may receive images similar to the images 500-800 and use them to identify patterns in the images associated with the condition of the surface of the DPF 122. In some embodiments, the images 500-800 may include corresponding descriptions detailing the condition of the surface of the DPF 122 in each respective image. The image recognition system 200 may receive these images and descriptions and use them to train the image recognition system 200 on pattern recognition, surface assessment, and accuracy.

[0122] As utilized herein, the terms “approximately,” “about,” “substantially”, and similar terms are intended to have a broad meaning in harmony with the common and accepted usage by those of ordinary skill in the art to which the subject matter of this disclosure pertains. It should be understood by those of skill in the art who review this disclosure that these terms are intended to allow a description of certain features described and claimed without restricting the scope of these features to the precise numerical ranges provided. Accordingly, these terms should be interpreted as indicating that insubstantial or inconsequential modifications or alterations of the subject matter described and claimed are considered to be within the scope of the disclosure as recited in the appended claims.[0123J It should be noted that the term “exemplary” and variations thereof, as used herein to describe various embodiments, are intended to indicate that such embodiments are possible examples, representations, or illustrations of possible embodiments (and such terms are not intended to connote that such embodiments are necessarily extraordinary or superlative examples).

[0124] The term “coupled” and variations thereof, as used herein, means the joining of two members directly or indirectly to one another. Such joining may be stationary (e.g., permanent or fixed) or moveable (e.g., removable or releasable). Such joining may beAtty. Dkt. No.: 106389-9772achieved with the two members coupled directly to each other, with the two members coupled to each other using one or more separate intervening members, or with the two members coupled to each other using an intervening member that is integrally formed as a single unitary body with one of the two members. If “coupled” or variations thereof are modified by an additional term (e.g., directly coupled), the generic definition of “coupled” provided above is modified by the plain language meaning of the additional term (e.g., “directly coupled” means the joining of two members without any separate intervening member), resulting in a narrower definition than the generic definition of “coupled” provided above. Such coupling may be mechanical, electrical, or fluidic. For example, circuit A communicably “coupled” to circuit B may signify that the circuit A communicates directly with circuit B (i.e., no intermediary) or communicates indirectly with circuit B (e.g., through one or more intermediaries).

[0125] References herein to the positions of elements (e.g., “top,” “bottom,” “above,” “below”) are merely used to describe the orientation of various elements in the FIGURES. It should be noted that the orientation of various elements may differ according to other exemplary embodiments, and that such variations are intended to be encompassed by the present disclosure.

[0126] While various circuits with particular functionality are shown in FIG. 2, it should be understood that the computing system 230 may include any number of circuits for completing the functions described herein. Additional circuits with additional functionality may also be included. Further, the computing system 230 may further control other activity beyond the scope of the present disclosure.[0127| As mentioned above and in one configuration, the “circuits” may be implemented in machine-readable medium for execution by one or more of various types of processors, such as the processor 246 of FIG. 2. Executable code may, for instance, comprise one or more physical or logical blocks of computer instructions, which may, for instance, be organized as an object, procedure, or function. Nevertheless, the executables need not be physically located together, but may comprise disparate instructions stored in different locations which, when joined logically together, comprise the circuit and achieve the stated purpose for the circuit. Indeed, a circuit of computer readable program code may be a single instruction, orAtty. Dkt. No.: 106389-9772many instructions, and may even be distributed over several different code segments, among different programs, and across several memory devices. Similarly, operational data may be identified and illustrated herein within circuits and may be embodied in any suitable form and organized within any suitable type of data structure. The operational data may be collected as a single data set or may be distributed over different locations including over different storage devices, and may exist, at least partially, merely as electronic signals on a system or network.[0128| While the term “processor” is briefly defined above, the term “processor” and “processing circuit” are meant to be broadly interpreted. In this regard and as mentioned above, the “processor” may be implemented as one or more processors, application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), digital signal processors (DSPs), or other suitable electronic data processing components structured to execute instructions provided by memory. The one or more processors may take the form of a single core processor, multi-core processor (e.g., a dual core processor, triple core processor, quad core processor, etc.), microprocessor, etc. In some embodiments, the one or more processors may be external to the apparatus, for example the one or more processors may be a remote processor (e.g., a cloud-based processor). Alternatively or additionally, the one or more processors may be internal and / or local to the apparatus. In this regard, a given circuit or components thereof may be disposed locally (e.g., as part of a local server, a local computing system, etc.) or remotely (e.g., as part of a remote server such as a cloud-based server). To that end, a “circuit” as described herein may include components that are distributed across one or more locations.[0129| Embodiments within the scope of the present disclosure include program products comprising computer or machine-readable media for carrying or having computer or machine-executable instructions or data structures stored thereon. Such machine-readable media can be any available media that can be accessed by a computer. The computer readable medium may be a tangible computer readable storage medium storing the computer readable program code. The computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, holographic, micromechanical, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the computer readable medium may include but are not limited toAtty. Dkt. No.: 106389-9772a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), a portable compact disc read-only memory (CD-ROM), a digital versatile disc (DVD), an optical storage device, a magnetic storage device, a holographic storage medium, a micromechanical storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, and / or store computer readable program code for use by and / or in connection with an instruction execution system, apparatus, or device. Machine-executable instructions include, for example, instructions and data which cause a computer or processing machine to perform a certain function or group of functions.

[0130] The computer readable medium may also be a computer readable signal medium. A computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated signal may take any of a variety of forms, including, but not limited to, electrical, electro-magnetic, magnetic, optical, or any suitable combination thereof. A computer readable signal medium may be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport computer readable program code for use by or in connection with an instruction execution system, apparatus, or device. Computer readable program code embodied on a computer readable signal medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, Radio Frequency (RF), or the like, or any suitable combination of the foregoing.

[0131] In one embodiment, the computer readable medium may comprise a combination of one or more computer readable storage mediums and one or more computer readable signal mediums. For example, computer readable program code may be both propagated as an electro-magnetic signal through a fiber optic cable for execution by a processor and stored on RAM storage device for execution by the processor.[01321 Computer readable program code for carrying out operations for aspects of the present disclosure may be written in any combination of one or more other programming languages, including an object-oriented programming language such as Java, Smalltalk, C++ or the likeAtty. Dkt. No.: 106389-9772and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The computer readable program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone computer-readable package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet Service Provider).10133] The program code may also be stored in a computer readable medium that can direct a computer, other programmable data processing apparatus, or other devices to function in a particular manner, such that the instructions stored in the computer readable medium produce an article of manufacture including instructions which implement the function / act specified in the schematic flowchart diagrams and / or schematic block diagrams block or blocks.

[0134] Although the figures and description may illustrate a specific order of method steps, the order of such steps may differ from what is depicted and described, unless specified differently above. Also, two or more steps may be performed concurrently or with partial concurrence, unless specified differently above. Such variation may depend, for example, on the software and hardware systems chosen and on designer choice. All such variations are within the scope of the disclosure. Likewise, software implementations of the described methods could be accomplished with standard programming techniques with rule-based logic and other logic to accomplish the various connection steps, processing steps, comparison steps, and decision steps.[01351 It is important to note that the construction and arrangement of the apparatus and system as shown in the various exemplary embodiments is illustrative only. Additionally, any element disclosed in one embodiment may be incorporated or utilized with any other embodiment disclosed herein.

Claims

Atty. Dkt. No.: 106389-9772WHAT IS CLAIMED IS:

1. A system for assessing a surface condition of a component of an aftertreatment system comprising:a computing system coupled to an image capture device, the computing system comprising one or more processors and one or more memory devices storing instructions therein that, when executed by the one or more processors, cause the one or more processors to perform operations comprising:receiving, from the image capture device, a first image of a surface of the component;extracting a first feature from the first image;identifying a second feature that corresponds to the first feature based on comparing the first feature to a set of historical features comprising at least the second feature;responsive to identifying the second feature, receiving a condition of the component, wherein the condition of the component corresponds to the second feature; and generating a notification for display on a user interface, the notification indicating the condition of the component.

2. The system of claim 1, wherein the component of the aftertreatment system is a particulate filter.

3. The system of claim 1, wherein the image capture device is a borescope.

4. The system of claim 1, wherein the instructions, when executed by the one or more processors, further cause the one or more processors to perform operations comprising processing the first image, wherein processing the first image comprises at least one of: removing at least one unidentified feature from the first image;removing noise from the first image; ormodifying a format of the first image.

5. The system of claim 1, wherein extracting the first feature from the first image comprises:Atty. Dkt. No.: 106389-9772parsing, using a machine learning model, a set of features from the first image that comprise at least the first feature; andclassifying, using the machine learning model, at least the first feature.

6. The system of claim 1, wherein the notification comprises at least one of a qualitative description regarding a deposit coverage on a surface of the component or an identification of a deposit type.

7. The system of claim 1, wherein the notification comprises one or more actionable recommendations regarding maintenance of the component, one or more descriptions regarding the condition, or one or more probable failure modes regarding the component.

8. The system of claim 7, wherein the one or more actionable recommendations comprise at least one of an indication that there is no maintenance required, an indication that the component needs to be replaced, or an indication that further testing is required.

9. A method for assessing a surface condition of a component of an aftertreatment system, the method comprising:receiving, by a computing system, a first image of a surface of the component of the aftertreatment system;extracting, by the computing system, visual data from the first image; classifying, by the computing system, the visual data based on at least one of a distribution pattern or a color contained in the visual data;generating, by the computing system, a report regarding a condition of the surface of the component;generating, by the computing system, a recommendation regarding maintenance of the aftertreatment system; andproviding, by the computing system, a maintenance recommendation to a user device, the maintenance recommendation comprising a notification indicating the condition of the component.Atty. Dkt. No.: 106389-977210. The method of claim 9, wherein the visual data comprises an indication regarding at least one of a deposit morphology on the surface, a deposit pattern on the surface, or a distribution of particulates forming abnormalities on the surface.

11. The method of claim 9, wherein classifying the visual data comprises comparing the first image to a plurality of stored images to identify at least one pattern or at least one abnormality in the visual data, and wherein the method further comprises:characterizing, by the computing system, the first image based on the at least one pattern or the at least one abnormality.

12. The method of claim 9, wherein the maintenance recommendation comprises at least one of an indication that there is no maintenance required, an indication that the component needs to be replaced, or an indication that further testing is required.

13. The method of claim 9, wherein the component is at least one of a particulate filter, an oxidation catalyst, or a selective catalytic reduction system.

14. The method of claim 9, further comprising processing the first image, wherein processing the first image comprises at least one of removing at least one unidentified feature from the first image, removing noise from the first image, or modifying a format of the first image.

15. The method of claim 9, wherein the visual data comprises a first feature of the first image.

16. The method of claim 15, further comprising:identifying a second feature that corresponds to the first feature based on comparing the first feature to a set of historical features comprising at least the second feature; and responsive to identifying the second feature, receiving a condition of the surface, wherein the condition corresponds to the second feature.

17. A system for generating maintenance recommendations based on images of an aftertreatment system or component thereof, the system comprising:Atty. Dkt. No.: 106389-9772a computing system comprising one or more processors and one or more memory devices storing instructions therein that, when executed by the one or more processors, cause the one or more processors to perform operations comprising:receiving an image of a component of the aftertreatment system; extracting visual data from the image of the component of the aftertreatment system;providing the visual data to a machine learning model configured to identify and classify information in the visual data;classifying, via the machine learning model, the visual data based on at least one of a distribution contained in the visual data, a color contained in the visual data, a pattern in the visual data, or an abnormality in the visual data, wherein the classification of the visual data includes comparing the visual data to stored historical visual data;characterizing, by the machine learning model, the image of the component based on the classification of the visual data;determining, by the machine learning model, a condition of the component; generating a report regarding the condition of the component of the aftertreatment system based on the determined condition; andoutputting one or more maintenance recommendations regarding the component of the aftertreatment system to a display device.

18. The system of claim 17, wherein the classification of the visual data further comprises:identifying a first feature of the visual data based on comparing the visual data to the stored historical visual data, the stored historical visual data comprising a second feature that corresponds to the first feature of the visual data; andbased on the first feature and the second feature, determining the condition of the component.

19. The system of claim 17, wherein the stored historical visual data comprises a set of historical features, the set of historical features comprising a plurality of features corresponding to the condition of the component.Atty. Dkt. No.: 106389-977220. The system of claim 17, wherein the image of the component of the aftertreatment system comprises one or more images of a surface of at least one of a particulate filter, an oxidation catalyst, or of a selective catalytic reduction system.

Images