Parking validation

Abstract

Systems and methods described herein are directed to incentive and reward validation. The systems and methods may receive data representing an image of evidence associated with a transaction between a user and a vendor, wherein the evidence is devoid of indicia specifically identifying a qualification for a parking discount; analyze the data to identify discernable information within the evidence, the discernable information including at least one of a vendor identifier, a transaction location, a timestamp, or a transaction value; generate validation qualifying information by processing the discernable information through a machine-learned evidence model; match the validation qualifying information against a dataset of discount offers to identify a prescribed parking discount; and cause to update a parking fee associated with a parking session of the user to include the prescribed parking discount.

Description

Attorney Docket No.: EXG-1PARKING VALIDATIONCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] The present application claims priority to U.S. Provisional Patent Application Serial No. 63 / 777,094 filed on March 25, 2025, the disclosure of which is incorporated by reference herein in its entirety.FIELD

[0002] The present disclosure relates generally to incentive and reward validation, and more particularly to physical-world cross-promotional validation, and more particularly to parking validation.BACKGROUND

[0003] Retail vendors, including consumer good vendors and hospitality retail vendors, such as department stores, big-box retailers, convenience stores, supermarkets, restaurants, and the like, often entice shoppers and diners through incentive programs. One ty pe of incentive program involves parking vouchers, whereby the shopper or diner receives a voucher that reduces the cost of parking in exchange for meeting a criterion, such as a dollar amount spent with the retail vendor.

[0004] Historically, parking vouchers were handled using a physical voucher system. For example, a worker at the retail vendor would manually veril whether the criterion was met by a customer and, upon verifying the criterion, issue a physical voucher that the customer could use to receive a discount when exiting the parking facility. The physical voucher traditionally included things like prepaid parking tickets, stickers that attach to existing tickets, validating tokens, and the like. These physical vouchers were redeemed through interaction with parking facility7equipment, such as parking payment kiosks, sometimes referred to as automated pay stations (APS), exit pay machines, parking ticket validators, or pay terminals.

[0005] While beneficial to attract customers, such voucher systems may increase checkout time by requiring the customer to perform additional steps at checkout. That is, in addition to inserting payment and the parking ticket, the customer must insert the physical voucher. The customer may additionally want to verify that the discountAttorney Docket No.: EXG-1was correctly applied, requiring yet further time to complete the transaction. Given that parking payment kiosks were historically located at egress points, it is common for lines of traffic to build where checkout time is long.

[0006] To solve this problem, parking facilities have introduced walk-up payment kiosks where parkers are able to process their parking tickets in advance, before approaching the egress point. Once at the egress point, the parker inserts a ticket or other voucher of payment and a barrier is opened to allow the parker to exit the facility. While reducing buildup of traffic at the egress points, it is common for these walk-up payment kiosks to be areas of danger since parkers must access their wallets to interact therewith and due to their exposure and location.

[0007] Accordingly, improved parking incentive programs and associated methods and systems are desired in the art. In particular, parking incentive programs, systems, and methods which allow parkers to redeem parking incentives without physical vouchers and without having to interact with parking access / revenue control systems (PARCS) hardware, such as gates, kiosks, walk-up or drive-up payment kiosks, etc. would be advantageous.BRIEF DESCRIPTION

[0008] Aspects and advantages of the invention in accordance with the present disclosure will be set forth in part in the following description, or may be obvious from the description, or may be learned through practice of the technology.

[0009] In accordance with one embodiment, a non-transitory computer-readable medium storing instructions which, when executed by one or more processors, causes performance of operations is provided. The operations include receiving, at a remote server, data representing an image of evidence associated with a transaction between a user and a vendor, wherein the evidence is devoid of indicia specifically identifying a qualification for a parking discount; analyzing, at the remote server, the data to identify discernable information within the evidence, the discernable information including at least one of a vendor identifier, a transaction location, a timestamp, or a transaction value; generating validation qualifying information by processing the discernable information through a machine-learned evidence model; matching the validation qualifying information against a dataset of discount offers to identify' aAttorney Docket No.: EXG-1prescribed parking discount; and causing to update a parking fee associated with a parking session of the user to include the prescribed parking discount.

[0010] In accordance with another embodiment, a parking validation system is provided. The parking validation system includes a communication interface configured to receive data representing an image of evidence associated with a transaction between a user and a retail vendor, wherein the evidence is devoid of indicia specifically identifying a qualification for a parking discount; a memory storing a model and a dataset of discount offers; and one or more processors in communication with the communication interface and the memory, the one or more processors configured to: identify discernable information within the data, the discernable information including at least one of a vendor identifier, a transaction location, a timestamp, or a transaction value; execute the model to generate validation qualifying information based on the discernable information; compare the validation qualifying information to the dataset of discount offers to identify a matched discount offer; and update a parking fee associated with a parking session of the user based on a prescribed parking discount associated with the matched discount offer.

[0011] In accordance with another embodiment, a method for validating a discount parking fee is provided. The method comprises receiving, at a remote server, data representing an image of evidence associated with a transaction between a user and a retail vendor, the evidence being devoid of specific indicia identifying a qualification for a parking discount; identifying, via one or more processors at the remote server, discernable information within the data, the discernable information including at least one of a vendor name, a transaction location, a timestamp, or a transaction value; determining validation qualifying information by processing the discernable information through a machine-learned evidence model; identifying a matched discount offer by comparing the validation qualifying information to a dataset of discount offers each associated with a prescribed parking discount; applying the prescribed parking discount to a parking fee associated with a parking session; and causing to initiate a transaction to effect payment of the parking fee.

[0012] These and other features, aspects and advantages of the present invention will become better understood with reference to the following description and appended claims. The accompanying drawings, which are incorporated in andAttorney Docket No.: EXG-1constitute a part of this specification, illustrate embodiments of the technology and, together with the description, serve to explain the principles of the technology.BRIEF DESCRIPTION OF THE DRAWINGS

[0013] A full and enabling disclosure of the present invention, including the best mode of making and using the present systems and methods, directed to one of ordinary skill in the art, is set forth in the specification, which makes reference to the appended figures, in which:

[0014] FIG. 1 is a schematic aerial view of a retail location including a plurality of retail vendors and a parking facility in accordance with embodiments of the present disclosure;

[0015] FIG. 2 is a flow chart of a method of validating a discount for a parking fee in accordance with embodiments of the present disclosure;

[0016] FIG. 3 is an example of evidence captured as part of a method of validating a discount for a parking fee in accordance with embodiments of the present disclosure;

[0017] FIG. 4 illustrates a schematic of a mobile device in accordance with embodiments of the present disclosure;

[0018] FIG. 5 illustrates a mobile client application in accordance with embodiments of the present disclosure;

[0019] FIG. 6 illustrates a schematic of a remote server in accordance with embodiments of the present disclosure; and

[0020] FIG. 7 illustrates a flow' diagram of a validation hub executing a crosssector promotional mapping in accordance with embodiments of the present disclosure.DETAILED DESCRIPTION

[0021] Reference now will be made in detail to embodiments of the present invention, one or more examples of which are illustrated in the drawings. The word “exemplary” is used herein to mean “serving as an example, instance, or illustration.” Any implementation described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other implementations. Moreover, eachAttorney Docket No.: EXG-1example is provided by way of explanation, rather than limitation of, the technology. In fact, it will be apparent to those skilled in the art that modifications and variations can be made in the present technology without departing from the scope or spirit of the claimed technology. For instance, features illustrated or described as part of one embodiment can be used with another embodiment to yield a still further embodiment. Thus, it is intended that the present disclosure covers such modifications and variations as come within the scope of the appended claims and their equivalents. The detailed description uses numerical and letter designations to refer to features in the drawings. Like or similar designations in the drawings and description have been used to refer to like or similar parts of the invention.

[0022] As used herein, the terms ''first' “second”, and "third" may be used interchangeably to distinguish one component from another and are not intended to signify location or importance of the individual components. The singular forms “a,” “an,” and “the” include plural references unless the context clearly dictates otherwise. The terms “coupled,” “fixed.” “attached to,” and the like refer to both direct coupling, fixing, or attaching, as well as indirect coupling, fixing, or attaching through one or more intermediate components or features, unless otherwise specified herein. As used herein, the terms “comprises,” “comprising.” “includes,” “including,” “has,” “having” or any other variation thereof are intended to cover a non-exclusive inclusion For example, a process, method, article, or apparatus that comprises a list of features is not necessarily limited only to those features but may include other features not expressly listed or inherent to such process, method, article, or apparatus. Further, unless expressly stated to the contrary, “or” refers to an inclusive- or and not to an exclusive- or. For example, a condition A or B is satisfied by any one of the following: A is true (or present) and B is false (or not present). A is false (or not present) and B is true (or present), and both A and B are true (or present).

[0023] Benefits, other advantages, and solutions to problems are described below with regard to specific embodiments. However, the benefits, advantages, solutions to problems, and any feature(s) that may cause any benefit, advantage, or solution to occur or become more pronounced are not to be construed as a critical, required, or essential feature of any or all the claims.Attorney Docket No.: EXG-1

[0024] The term retail vendor is used herein to refer to a wide variety of businesses and is not intended to be limited to any particular type of business. Retail vendors can include, for example, general retail vendors like department stores, specialty shops, big-box retailers, and the like. Retail vendors can also include food and beverage retailers like grocery' stores, supermarkets, bakeries, beverage shops, specialty food stores, and the like. Retail vendors can also include restaurants and food service vendors like full-service restaurants, fast food, cafes, catering businesses, and the like. Retail vendors can also include wholesale food supplies like bulk food distributors for restaurants and grocery stores. Retail vendors can also include convenience stores and gas stations. Retail vendors can also include service providers, such as professional businesses like architects, law firms, banks, accountants, and the like. Retail vendors can also include housing providers, such as apartment buildings, community' buildings, and the like. Retail vendors can also include community sendees, such as public works, gyms, health clinics, hospitals, and the like. Retail vendors can also include commercial owners that operate a commercial space and lease the commercial space to one or more businesses. The term retail vendor is intended generally to refer to a business that may offer an incentive program to discount parking for a parked vehicle, such as for a customer frequenting the business.

[0025] The term parking facility is intended to refer to a designated area, structure, or space where one or more vehicles can be parked. The parking facility may include an open surface lot, a covered (e.g., multi-level) garage, an underground structure, or the like. The parking facility can be privately or publicly operated. The parking facility may charge a fee for parking. In some instances, the parking fee is fixed (e.g., hourly). In other instances, the parking fee is based on one or more variables, such as duration of a parking session, vehicle type, time of day, day of week, congestion or remaining availability' (sometimes referred to as surge rates), or the like.

[0026] In one embodiment, the parking facility' includes systems for managing vehicle entry, exit, and space allocation. For example, the parking facility can include one or more parking access / revenue control systems (PARCS), such as for example, movable barriers, e.g., garage doors or gate arms, that are operably driven by barrier operators between open and closed positions. The parking facility can further includeAttorney Docket No.: EXG-1other PARCS hardware, such as payment kiosks and / or ticketing systems. In some instances, the parking facility’ may not include PARCS hardware. For example, the parking facility may be ungated. Ungated parking facilities may rely, for example, on image footage, as captured using an onsite and / or nearby image capture device (e.g., a camera), to monitor vehicle ingress and / or egress. The camera may be linked to a logic device which verifies whether a parking fee was paid. Where no parking fee w as paid, the logic device, or another system in communication therewith, may issue a ticket, optionally including a penalty, for payment of the parking fee. The ticket price may be automatically deducted from an online bank account, mailed to the driver / owner of the vehicle, used in a follow up ticketing event (e.g., when the vehicle is again detected at the parking facility, or the like). Yet alternatively, payment may occur in response to vehicle egress from the parking facility without any action taken by the parker prior to exiting. For example, the driver may have payment information on file which is automatically debited in response to the vehicle exiting the parking facility.

[0027] In some implementations the parking facility is owned by the retail vendor, or a combination of retail vendors. The parking facility may be further operated by the retail vendor. For example, a residential apartment building may ow n and operate a nearby parking facility. Operating the parking facility refers to managing and controlling functions associated with the parking facility, including vehicle entry and exit, space allocation, fee collection, security enforcement, maintenance, and the like. This may involve overseeing payment systems, monitoring occupancy levels, and implementing access controls. Often, the party owning the parking facility contracts with a management company to operate the parking facility.

[0028] In some implementations, the parking facility is owned by an owner of the retail space to which one or more retail vendors lease space. In this case, the parking facility is owned by a non-retail vendor. The non-retail vendor typically contracts with a management company to operate the parking facility.

[0029] In certain embodiments, the parking facility may instead include another type of facility, such as an event venue (e.g., a sports stadium, a music theater, a movie theater, a public works location, etc.).Attorney Docket No.: EXG-1

[0030] Unless described to the contrary7, the terms customer and parker are used herein to describe a person, or people, that have a parking session and seeking a discounted parking fee based on meeting a criterion set by a retail vendor.

[0031] In general, embodiments described herein are directed to validation of parking, and more particularly to application of parking fee discounts based on presentation of evidence supporting qualification for an available discount offer. Implementations described herein allow a customer (sometimes referred to as a user or a parker) to capture evidence of a discount-qualifying transaction to receive partial or full validation of a parking fee. Notably, the captured evidence is, or may be, free (i.e., devoid) of indicia supporting qualification for the discount offer. That is, the evidence does not contain indicia supporting validation, such as text stating ‘QUALIFIED FOR DISCOUNT” or “FREE PARKING”, or a barcode or QR code including or referencing information that provides the validation, or the like. Such evidence was traditionally added to receipts or parking vouchers by the retailer on a transaction-by-transaction basis. Such action required the retailer to maintain onsite equipment and / or costly systems to affect the evidence and ultimately resulted in the retailer having to spend time and resources to validate customer parking fee discounts. As contemplated herein, the evidence includes standard transaction information, such as for example, price, time, retail vendor information, and the like, that is individually or collectively used to determine whether the evidence warrants discounted parking at least partially in view of one or more available discount offers.

[0032] The evidence is captured using a mobile device, such as a smart phone having an integrated image capture device. The evidence is transmitted to a remote server where qualifying information is compared against available discount offers. Alternatively, the evidence may be examined locally at the user's mobile device. Where qualification occurs, i.e., the qualifying information matches a criterion associated with an available discount offer, the associated discount is applied to the parking fee and payment of the discounted parking fee is affected. The customer is then able to exit the parking facility without having to interact with any physical payment kiosks, such as drive-up payment kiosks, walk-up payment kiosks, or the like. No physical token, voucher, or sticker is provided. In some instances, the customer can fully complete the validation process prior to arrival at the parkingAttorney Docket No.: EXG-1facility, such as while in transit to the parking facility from the retail vendor, from inside the safety and convenience of the retail vendor's facility, or from another location remote from the parking facility. In this regard, the customer is not slowed by any local processing lag and can perform the transaction without having to spend time at the parking facility, thereby increasing customer satisfaction and safety’.

[0033] A key technical aspect of the systems and methods described herein is the utilization of a machine-learned evidence model to perform high-level inference on unstructured transaction data. Unlike traditional systems that require specific, preformatted codes, the present system identifies qualifying information from standard identifying blocks (e g., vendor names, itemized values, and timestamps, etc.) that are typically included on transaction-confirming documents for non-parking purposes. The machine-learned evidence model is trained on diverse datasets to recognize patterns and correlations that distinguish qualifying transactions from non-qualifying transactions, even in the presence of visual noise, surface distortions, or variable document layouts.

[0034] In addition to automated validation, the system may employ one or more generative models, such as a large language model (LLM), to facilitate real-time, context-aware interaction with the user. These models are configured to provide informative notifications regarding the status of the validation process, explain the parameters of identified discount offers, and / or provide specific troubleshooting instructions if the captured evidence is insufficient for processing. By generating natural language responses and interactive prompts, the system reduces user cognitive load and ensures a high success rate for the validation process without requiring human intervention.

[0035] To increase data security and achieve regulatory compliance, the system can include a privacy -preserving layer that obfuscates personally identifiable information (PII) at the edge. Captured data may be randomized, hashed, or encoded using techniques, such as Base64 encoding, e.g., prior to transmission to a remote server, ensuring that sensitive user financial information is never stored in a vulnerable format. This distributed processing approach allows the system to verity' transaction qualification and affect settlement while maintaining a zero-trust environment regarding the user's private transaction history.Attorney Docket No.: EXG-1

[0036] In some embodiments, the systems and methods are configured to ingest and process arbitrary evidence, which refers to any verifiable physical or digital artifact representing a real-world event or transaction. While a sales receipt is an example of such evidence, the arbitrary nature of the evidence allows the system to identify qualifying information from a wide array of sources, including but not limited to event tickets, digital boarding passes, physical tokens, collectible cards, membership cards, repair orders, or the like. The machine-learned evidence model is configured to perform semantic analysis on these varied formats to extract a proof of presence or proof of purchase that serves as the trigger for a downstream incentive.

[0037] The system may function as a cross-promotional validation hub, mapping evidence from a first commercial sector, such as a sporting event or hospitality vendor, to a discount or incentive in a second, unrelated commercial sector, such as a retail sporting goods store or a parking facility'. For example, the system may analyze a ticket stub from an athletic event to identify' the event location and date; upon verification, the system may automatically update a user’s profile to include a discount at a participating local retailer. This can enable cross-promotional service(s) for the physical world, where the act of presenting evidence at a digital gateway unlocks distributed value across a network of participating vendors.

[0038] The system may utilize an SMS / MMS-based communication interface. In this implementation, a user may capture an image of the arbitrary evidence and transmit the image (or data associated with the image) via a standard messaging protocol to a designated shortcode or long-code phone number. The remote server is configured to identify' the user session via the originating mobile identifier, parse the image data, and return a responsive link or natural language notification containing the available incentives. This allows for validation without requiring the user to download a native application prior to the first transaction. The communication interface can also, or alternatively, be via a native application instantiated, e.g., on the mobile device.

[0039] In certain embodiments, the system may incorporate a physical-digital linking mechanism to tie a user’s physical presence to a specific digital validation session. This may involve the use of a unique physical token, such as an RFID card, a collectible card, or a wearable device, which is presented alongside the evidenceAttorney Docket No.: EXG-1during the image capture process. The machine-learned model may be trained to recognize the physical token within the frame of the image, utilizing its unique visual or electronic signature to associate the transaction with a specific user profile or vehicle, thereby providing an additional layer of multi-factor authentication for incentives.

[0040] The dataset of discount offers may be stored within a database. The database can track the temporal relevance of validated evidence. In this configuration, the value or availability of an incentive may be dynamically adjusted based on the time elapsed since the event represented by the evidence. For example, a validated restaurant receipt may trigger a full parking subsidy if redeemed within a first threshold time, a partial subsidy if redeemed within a second threshold time, and a transition to a retail-only discount if redeemed at a later date. Such temporal decay encourages consumer behavior while maintaining a persistent record of the user’s real-world activity7.

[0041] The persistent record of the user's real-world activity may be utilized by the one or more processors to establish a historical behavior profile, allowing the system to refine the machine-learned evidence model based on longitudinal user engagement. By analyzing the rate at which a user responds to varied temporal decay curves, the system can dynamically optimize both the initial value and the subsequent rate of reduction for future discount offers to maximize retail conversion and / or manage parking facility occupancy. Furthermore, this persistent record of the user’s real-world activity may serve as a foundational dataset for predictive modeling, wherein the system anticipates a user’s future presence at a parking facility or retail location based on past validation patterns and preemptively prepares associated cross-promotional incentives. Additionally, the predictive modeling may be configured to generate pre-validated incentives that are pushed to the user’s mobile device upon the system detecting the vehicle's ingress into the parking facility7, thereby creating a proactive, circular economy between the parking infrastructure and the retail vendors.

[0042] The system may use a generative multi-modal model to assist users in the event of a validation failure. If the machine-learned evidence model determines that the captured image is of insufficient quality, such as due to blur, lighting, or occlusions, the generative model may produce a specific, natural language instructionAttorney Docket No.: EXG-1or a visual overlay indicating the exact area of the document that needs to be rescanned. For instance, the system may generate a message indicating that the vendor address is cut off and provide instructions to move the camera, thereby providing a conversational experience for technical data capture.

[0043] An analysis engine may be further configured to provide contextual advertisement injection based on the identified evidence or data associated therewith. If the evidence indicates a purchase at a specific retail vendor, the generative model may not only provide a parking discount but also suggest offers from competing or complementary vendors in the same geographical vicinity.

[0044] To protect participating vendors, the system may incorporate automated fraud detection layers within the analysis engine. For example, the system may check the unique identifier of the evidence, such as a receipt serial number, against a previously redeemed ledger to prevent duplicate validation attempts. Furthermore, the system may utilize temporal-geospatial correlation to ensure that the time and location of the transaction are logically consistent with the current parking session or the user’s reported location, effectively identifying evidence that appears to be sourced from outside a permitted incentive zone.

[0045] The system may provide a vendor management interface that allows unrelated businesses to create cross-promotional networks. Through the vendor management interface, a first vendor can authorize a second vendor to accept its tickets as valid evidence for a discount. The vendor management interface can allow for the setting of budget caps, redemption windows, and specific qualifying events, which dynamically triggers a discount across the network of participating retail and parking facilities.

[0046] The vendor management interface may be configured to provide a suite of administrative controls that allow a retail vendor or a third-party administrator to define the fiscal and temporal parameters of an incentive program. For example, the interface may receive a budget cap representing a maximum financial liability' or a maximum number of individual redemptions permitted for a specific discount offer. Upon the one or more processors determining that the cumulative value of applied discounts has reached the budget cap, the system may be configured to automaticallyAttorney Docket No.: EXG-1transition the discount offer to an inactive state within the dataset, thereby preventing further redemptions without manual intervention.

[0047] The vendor management interface may facilitate the definition of redemption windows, which specify the precise dates and / or times during which a discount offer is valid. These windows may be static (e.g., every Tuesday from 4:00 PM to 8:00 PM) or dynamic (e.g., valid for three hours following the conclusion of a specific event). In some embodiments, the redemption window may be linked to memory, where the system analyzes the timestamp of the arbitrary evidence to ensure the evidence was generated within a permitted timeframe relative to the current parking session or the current time.

[0048] Specific qualifying events may be defined via the vendor management interface to serve as the trigger for cross-sector incentives. These events may include commercial transactions, such as a purchase exceeding a transaction value threshold at a restaurant, or non-commercial events, such as attendance at a sporting event, a team victory, or reaching a specific milestone in a loyalty program. The system may be configured to monitor external data feeds to verify the occurrence of a qualifying event (e.g., a "victory discount" triggered only if the local team wins), which then dynamically activates corresponding discount offers across the network of participating facilities. The network of participating facilities may include a plurality of unrelated commercial entities, such as parking garages, big-box retailers, and hospitality vendors, that are linked through the vendor management interface. When a qualifying event is identified from the evidence, the system may simultaneously update the parking fees for a user's current parking session while also generating a digital token for use at a neighboring retail location. This dynamic triggering allows for the creation of a cross-promotional web where a single piece of evidence, such as a stadium ticket, serves as a universal key to unlock multiple disparate benefits throughout a geographical region.

[0049] The vendor management interface may also provide real-time visibility into the performance of these triggers, allowing vendors to observe user engagement metrics and adjust qualifying criteria on-the-fly to optimize traffic flow or retail conversion. For instance, if a parking facility is nearing capacity, a vendor may use the interface to temporarily increase the "transaction value" requirement for a parkingAttorney Docket No.: EXG-1subsidy, thereby managing demand while still honoring high-value customer interactions.

[0050] The systems and methods described herein may incorporate an automated optimization engine that utilizes the historical behavior profile to implement predictive load balancing between retail foot traffic and parking availability. By leveraging real-time data feeds from the parking spot identification system and retail point-of-sale systems, the analysis engine can suggest or autonomously execute adjustments to incentive parameters (e.g., the transaction value or redemption window) without requiring manual oversight. For example, during a high-volume retail event where conversion is the primary KPI, the system may pivot to a yieldmanagement strategy, dynamically raising the validation threshold to prioritize premium parking spaces for customers with higher itemized transaction values while simultaneously pushing lower-tier spillover incentives to nearby, under-utilized parking facilities within the network. This creates a highly responsive, living commercial ecosystem where a digital validation hub acts as a physical-world traffic controller, ensuring that the allocation of limited infrastructure is always algorithmically aligned with the most valuable commercial outcomes for the participating vendor network.

[0051] In some implementations, the systems and methods described herein may utilize a prioritization algorithm within the analysis engine to determine the relevance of available discount offers based on the historical behavior profile. For example, if a user's engagement indicates a preference for specific retail categories, such as sporting goods over hospitality7, the generative model may prioritize the display of associated incentives. This relevance scoring may incorporate real-time variables such as a user’s current geospatial proximity to a participating facility or the real-time availability7of spaces within a parking facility' to ensure that the generated update provides the highest immediate utility7to the user.

[0052] The systems and methods described herein may extend the validation and incentive engine to non-parking services, such as electric vehicle (EV) charging, car washing services, micro-mobility rentals (e g., e-scooters or e-bikes), or the like. For instance, arbitrary evidence representing a transaction at a retail vendor may be used to provide a full or partial subsidy for EV charging fees incurred at a charging stationAttorney Docket No.: EXG-1located within or near the parking facility7. The update to the service fee may occur dynamically via the remote server's communication with the service infrastructure, allowing for the same frictionless, kiosk-free experience as provided for parking fees.

[0053] In instances where the arbitrary7evidence is a digital artifact, such as a digital ticket or a mobile wallet pass, the system may utilize a secure API integration or a cryptographic handshake to verify the authenticity7of the artifact. For example, the remote server may communicate with a third-party event management database to confirm that a unique ticket identifier is valid and has not been previously utilized for a validation event. Alternatively, the machine-learned evidence model may perform visual verification of security7features within a screenshot of the digital evidence, such as dynamic watermarks, rolling time codes, or specific UI elements unique to a verified vendor application.

[0054] In certain embodiments, the capture of arbitrary evidence is performed by7a specific, hardware-triggered physical action performed by the user to transform a transient optical state of the display into a persistent, electronically verifiable data artifact. This physical action, such as a multi-button press, a specific haptic gesture, or a biometric-authenticated command, e.g., via the user input of a mobile device, initiates a low-level operating system routine that freezes the current frame buffer of the mobile device. This process constitutes a technical transformation of a temporary visual arrangement of pixels (originally intended for human consumption) into a fixed, pixel-based evidence payload that is algorithmically distinct from a backend data transfer or a passive API handshake. The resulting screenshot can sen e as a durable surrogate for the physical-world transaction, carry ing unique visual metadata (such as system clock overlays, application-specific UI signatures, and dynamic security7watermarks, etc.) that the validation hub utilizes to verify the authenticity of the digital evidence in a manner that mimics the forensic analysis of a physical receipt.

[0055] To prevent fraudulent replication of physical tokens, such as collectible cards or RFID-enabled objects, the machine-learned evidence model may be configured to detect anti-spoofing features during the image capture process. These features may include holographic elements, variable-depth textures, or specific reflectance patterns that distinguish a physical token from a digital reproductionAttorney Docket No.: EXG-1displayed on a screen. The system may further utilize the image capture device's flash or ambient light sensors to verify the presence of these physical characteristics before associating the transaction with a specific user profile or vehicle.

[0056] The machine-learned evidence model may be configured for continuous learning, wherein the results of each validation event are fed back into a training pipeline to refine the model's accuracy. If the generative model identifies a recurring failure in document edge detection or character recognition for a specific vendor's unique receipt format, the system may automatically prioritize the labeling of similar data samples to update the ground truth labels. This iterative refinement allows the system to adapt to changing document layouts and evolving retail formats without manual reconfiguration of the underlying logic, maintaining the high success rate of the automated validation process.

[0057] Embodiments of the present disclosure provide a technical platform for cross-sector incentive validation that transforms unstructured commercial artifacts into functional digital keys. By utilizing a remote server including a data processing module and a high-throughput analysis engine, the system employs a machine-learned evidence model to derive validation qualifying information from arbitrary evidence (such as a physical receipt or a hardware-triggered screenshot of a digital ticket) that is inherently devoid of specific parking validation indicia. This automated verification process executes a multimodal temporal-geospatial correlation between the evidence metadata and the real-time physical state of a service facility, such as a parking garage or EV charging station, to trigger a secure hardware handshake. This handshake culminates in the transmission of a control signal to an onboard actuator of a movable barrier operator or service interface, facilitating frictionless facility egress or service activation without requiring a physical voucher or interaction with a traditional payment kiosk. Effectively functioning as a distributed cross-promotional hub, the system can algorithmically map qualifying events from a first commercial sector to dynamic, time-decaying incentives in a second, unrelated commercial sector, thereby optimizing infrastructure throughput and retail conversion through a non-passive, verifiable digital gateway. These and other advantages will be further described with respect to the drawings.Attorney Docket No.: EXG-1

[0058] Referring now to the drawings, FIG. 1 illustrates a schematic aerial view of an exemplary retail location 100 including a plurality of retail vendors 102 and a parking facility 104 located near the retail vendors 102.

[0059] In some implementations, the retail vendors 102 are located adjacent to the parking facility 104, such as immediately adjacent therewith. In other implementations, the retail vendors 102 can be separated from the parking facility' 104, such as for example, when the parking facility is a publicly- or privately-operated multi-tier garage servicing a greater local area in the vicinity’ of the retail vendor 102. In some embodiments, the retail location 100 may further include an alternative evidence source, such as a sports arena, stadium, or event venue, located remotely from the retail vendors 102 and / or the parking facility 104.

[0060] The depicted parking facility 104 is a surface lot 106 defining a plurality of individual parking spaces 108. The surface lot 106 may alternatively be a covered lot, a multi-level garage, a subterranean parking location, a below-retail garage, or the like. One or more of the individual parking spaces 108 may be equipped with an electric vehicle (EV) charging station 109. permitting the system to validate and update service fees for EV charging. Access to the parking facility 104 is restricted by a movable barrier 110 disposed at or near a perimeter 112 of the surface lot 106. The movable barrier 110 can include, for example, a barrier gate arm 114 that is driven by a motor between a raised position, whereby access to the surface lot 106 is granted, and a lowered position, whereby access to the surface lot 106 is restricted.

[0061] A drive-up payment kiosk 116 can be disposed at a location nearby the movable barrier 110 or contained at least partially within a housing of the movable barrier operator. The drive-up payment kiosk 116 can include a user interface configured to facilitate payment transactions through multiple methods, including cash, credit / debit cards, and mobile payment systems. The drive-up payment kiosk 116 can further include a touchscreen display, a ticket scanning mechanism, and / or a real-time transaction processing unit to optimize payment efficiency and reduce user wait times. The drive-up payment kiosk 116 can integrate with parking management infrastructure, enabling features such as receipt generation, contactless payment processing, and license plate recognition for automated entry and exit control. When entering the parking facility' 104, a driver may be provided with a ticket from theAttorney Docket No.: EXG-fdrive-up payment kiosk 116. The driver may retain the ticket to present to the drive-up payment kiosk 116 upon departure from the parking facility 104.

[0062] In some implementations, the parking facility 104 can further, or alternatively, include a walk-up payment kiosk 118. The walk-up payment kiosk 118 is typically located at a pedestrian-accessible area spaced apart from the movable barrier 110. The walk-up payment kiosk 118 allows for payment of parking fees prior to arrival at the movable barrier 110. The walk-up payment kiosk 118 can include any number of similar features as compared to the drive-up payment kiosk 116. For example, the walk-up payment kiosk 118 can include a user interface configured to facilitate payment transactions through multiple methods, including cash, credit / debit cards, and mobile payment systems. The walk-up payment kiosk 118 can further include a touchscreen display, a ticket scanning mechanism, and / or a real-time transaction processing unit. A vendor management terminal 119 may be disposed within one or more of the retail vendors 102 or at a remote administrator location, providing a multi-tenant interface for vendors to independently manage budget caps, redemption windows, and qualifying event triggers. Upon successful payment, the walk-up payment kiosk 118 (or a system in communication therewith) associates the transaction with a license plate of the vehicle, a validated ticket, or other identify ing information, thereby allowing the vehicle to exit seamlessly without delay at the movable barrier 110. This eliminates the need for in-vehicle payment processing at the exit gate, reducing congestion and improving overall traffic flow within the parking facility' 104.

[0063] The parking facility 104 can further include one or more cameras, such as a camera 120, for generating visual feed of the parking facility 104. vehicles 122 entering and / or exiting the parking facility 104, pedestrians moving through the parking facility’ 104, and the like. The camera 120 may be static or movable. In some implementations, the camera 120 can be mounted within the parking facility 104, such as mounted to a utility pole. In other implementations, the camera 120 is mounted external to the parking facility 104, such as to an adjacent building or structure. The exterior camera 120 can be pointed towards the parking facility 104 to capture information associated with vehicles 122 entering and exiting the parking facility 104. In some instances, the camera 120 can capture video feed of vehicle identifyingAttorney Docket No.: EXG-1information like license plate numbers. In some implementations, the camera feed captured by the camera 120 can be used to validate parking, generate security footage, troubleshoot problems, and the like.

[0064] The camera 120 can be in communication with a processor 128 and, optionally, transmit captured video feed to the processor 128. The processor 128 may be local, i.e., at the parking facility 104, or remote, e.g., at a remote server 124 as depicted in FIG. 1 through a network 126. such as a wireless network like the Internet. The processor 128 can be any suitable processing device (e.g., a control circuitry, a processor core, a microprocessor, an application specific integrated circuit, a field programmable gate array, a controller, a microcontroller, etc.) and can be one processor or a plurality of processors that are operatively connected. The processor(s) 128 may be coupled to memory 130. The memory 130 can include one or more non-transitory computer-readable storage media, such as RAM, ROM, EEPROM, EPROM, one or more memory devices, flash memory devices, etc., and combinations thereof. The memory 130 can store information that can be accessed by the processor(s) 128. For instance, the memory 130 (e.g., one or more non-transitory computer-readable storage mediums, memory devices) can include computer-readable instructions 132 that can be executed by the processor(s) 128. The instructions 132 can be software, firmware, or both written in any suitable programming language or can be implemented in firmware or hardware. Additionally, or alternatively, the instructions 132 can be executed in logically and / or virtually separate threads on processor(s). For example, the memory 130 can store instructions 132 that when executed by the processor(s) 128 cause the processor(s) 128 to perform operations such as any of the operations and functions as described herein.

[0065] The parking facility- 104 may further include a guidance system, such as for example, a parking spot identification system 134. The parking spot identification system 134 can include a network of sensors and indicator lights to guide drivers to available parking spaces 108. Each individual parking space 108 can be equipped with a sensor, such as an ultrasonic, infrared, or camera-based detection unit, which continuously monitors the presence or absence of a vehicle. When a parking space 108 is unoccupied, the sensor communicates this status to a central processing unit, which then activates an overhead or adjacent indicator light — typically green toAttorney Docket No.: EXG-1signify availability. Conversely, when a vehicle 122 occupies the parking space 108, the sensor detects vehicle presence and signals the control system to switch the indicator light to red or another designated color representing occupancy. The parking spot identification system 134 may further integrate with a centralized display or mobile application to provide real-time availability updates, enhancing parking efficiency and reducing congestion. Additionally, advanced implementations may incorporate machine learning algorithms to predict space availability trends based on historical usage patterns, further optimizing traffic flow within parking facilities.

[0066] The aforementioned systems and apparatuses, alone or in combination with other equipment at the parking facility 104, can enhance ease of parking and facilitate a smoother parking experience. This is particularly important for busy¬ parking lots where parking is tight and congested, such as when the neighboring retail vendors 102 are having big sales or other crowd-drawing events.

[0067] One method for retail vendors 102 to increase parking activity is byvalidating parking fees, also known as validated parking. Validated parking refers to a system in which a parking fee associated with parking at the parking facility 104 is partially or fully subsidized by a participating retail vendor 102, e.g., based on a criterion, such as a customer's qualifying transaction.

[0068] The parking facility 104 may be integrated with a validation mechanism such as described herein that allows retail vendors 102 to issue discounts upon a customer meeting one or more predetermined criterion. Such validation has been historically provided in the form of a physical stamp or ticket, a QR code, or a barcode that is presented at the kiosk 116, 118. The customer was required to take the physical evidence provided by the retail vendor 102 to the kiosk 116, 118 to complete the validation to receive the prescribed parking discount. In implementations described herein, such validation is replaced by an automated evidence-based digital validation technique that does not require special action by the retail vendor 102, other than initially setting up the discount and associated predetermined qualification criterion. Moreover, payment of the parking fee using the validation mechanism described herein may be made without requiring the customer to use a physical payment system, such as the drive-up payment kiosk 116 or even the walk-up payment kiosk 118. Instead, validation is completed using a mobile device. InAttorney Docket No.: EXG-1addition to reducing the burden on the retail vendor 102 and customer, embodiments described herein reduce the risk associated with customer dwelling time at the kiosk 116, 118 and allow the customer to quickly and efficiently return to their vehicle 122 and exit the parking facility without having to interact with unwanted equipment or people.

[0069] In one embodiment, the remote server 124 includes a data processing module and an analysis engine, each implemented in hardware, software, or a combination thereof to facilitate the high-throughput validation of arbitrary evidence. The data representing the image received at step 202 is first passed to the data processing module, which is configured to parse the data payload and extract relevant parameters associated with the user session and the evidence. The data processing module performs one or more normalization and / or transformation operations to convert the unstructured image data into a compatible format that ensures consistency with internal data schemas and expected communication protocols. This transformation process may include a software service executing within the server environment that is configured to analyze, interpret, or otherwise utilize the received data to perform application-specific functions before the data is passed to the analysis engine. The normalized data is then transmitted to the analysis engine, which comprises the one or more rule-based logic components, machine-learned evidence models, and / or statistical algorithms configured to evaluate the data according to the predefined retail or parking criteria. By partitioning the hardware and software in this manner, the system achieves a scalable and secure pipeline for processing diverse evidence formats while maintaining a high degree of technical accuracy in the subsequent validation steps.

[0070] FIG. 2 is a flowchart of a method 200 of validating a discount for a parking fee in accordance with an embodiment. In general, the method 200 will be described with reference to a parking environment including the retail vendors 102 and parking facility' 104 as described, for example, with respect to FIG. 1. However, other parking environments are contemplated herein. Although FIG. 2 depicts steps performed in a particular order for purposes of illustration and discussion, the method discussed herein is not limited to any particular order or arrangement. One skilled in the art, using the disclosure provided herein, will appreciate that various steps of theAttorney Docket No.: EXG-1method disclosed herein can be omitted, rearranged, combined, and / or adapted in various ways without deviating from the scope of the present disclosure.

[0071] The method 200 may include, at step 202, receiving, at a remote server (e.g., remote server 124 in FIG. 1), data representing an image of evidence associated with a transaction between a user and a retail vendor. As previously described, the evidence may comprise arbitrary evidence representing any verifiable physical or digital artifact of a real-world event or transaction, including but not limited to a sales receipt, an event ticket, a digital boarding pass, an order confirmation, or a repair order. In some embodiments, the image may further include a physical token, such as a collectible card or RFID-enabled object, presented alongside the transaction document to associate the physical presence of the user with the digital validation session.

[0072] Notably, the evidence is devoid of specific indicia identifying a qualification for a parking discount. That is, the evidence does not contain traditional validation markers such as a specific "FREE PARKING" stamp, a parking-specific QR code, or a barcode generated for the purpose of parking fee reduction. Instead, the systems and methods described herein rely on identifying qualifying information from standard identifying blocks typically included on transaction documents for nonparking purposes, such as vendor name, location, and timestamp. This lack of specialized indicia allows the system to utilize arbitrary evidence that was originally generated for a primary purpose unrelated to the parking validation process, such as for tax documentation, customer record-keeping, or order fulfillment. Traditional validation systems often require a retail vendor to maintain onsite equipment to append specific validation codes or physical markers to a customer's receipt, a process that can increase checkout duration and require significant retailer resources. By utilizing evidence devoid of such markers, the present system removes the requirement for the retail vendor to perform any affirmative act of validation at the point of sale, thereby decoupling the retail transaction from the parking fee adjustment. The systems and methods described herein effectively transform a standard commercial artifact into a functional digital key through server-side analysis. This shift ensures that the logic governing the discount (such as budget caps or redemption windows) resides within the remote server and the vendor managementAttorney Docket No.: EXG-1interface rather than being static on the face of a physical voucher. Consequently, the retail vendor can dynamically update their incentive programs in the dataset without needing to modify their physical checkout hardware or the format of their issued transaction documents.

[0073] In some implementations, the data is received via an SMS / MMS-based communication interface, wherein the user transmits the image via a standard messaging protocol to a designated shortcode or long-code phone number.Alternatively, the data may be received via a mobile client application executed on a mobile device (e.g., mobile device 500). Prior to or during transmission, the data may be processed through a privacy -preserving layer to obfuscate or encode personally identifiable information (PII) using techniques such as Base64 encoding or hashing to maintain a zero-trust environment regarding the user's private transaction history.

[0074] In other embodiments, the communication interface may be provided via a native application or mobile client application executed on the mobile device. The mobile client application may utilize the display module to present a graphical user interface on the display, allowing the user to interact with the evidence module to trigger the image capture device. This interface may provide real-time framing guidance, such as bounding boxes or reference markers, and utilize automated routines for document edge detection, perspective correction, and autofocusing to ensure the captured image meets the quality thresholds required by the machine-learned evidence model. The data may also, or alternatively, be received through other digital means, such as a web-based digital gateway, a browser extension, or a direct API integration with a third-party retail or event application. For instance, the system may be configured to monitor for the generation of digital evidence, such as an e-receipt or a digital ticket, and automatically pull the relevant data into the validation engine.

[0075] At step 204, the method 200 includes identifying, via one or more processors (e g., processor(s) 128) at the remote server, discernable information within the data. The discernable information includes at least one of a vendor name, a transaction location, a timestamp, or a transaction value. Identifying the discernable information may involve performing document edge detection, perspective correction, and optical character recognition (OCR) on the image data. In certain embodiments, aAttorney Docket No.: EXG-1generative model, such as a large language model (LLM), may be utilized to assist in parsing unstructured text or to generate a natural language query to the user if specific discernable information is occluded or missing.

[0076] At step 206, the method 200 includes determining validation qualifying information by processing the discernable information through a machine-learned evidence model. The machine-learned evidence model may perform high-level inference to extract a proof of purchase or proof of presence from the unstructured transaction data. For example, the model may be trained on diverse datasets to recognize patterns and correlations that distinguish qualifying transactions from nonqualifying ones, even in the presence of visual noise, surface distortions (e.g., wrinkles or creases), or variable document layouts.

[0077] In some embodiments, the model may further perform a multimodal temporal-geospatial correlation to ensure the transaction location and time are consistent w ith the user's current parking session. The multimodal temporal-geospatial correlation can function as a high-resolution technical synchronization between the metadata extracted from the arbitrary evidence and the real-time metadata associated with the active parking session. For instance, the analysis engine may cross-reference the specific vendor location and the timestamp identified on the evidence with the known geographical coordinates of the parking facility and the recorded ingress time of the vehicle. This correlation ensures that the evidence was generated within a logical proximity of the parking facility and during the active duration of the current parking session. Furthermore, the system may utilize real-time signals from the parking spot identification system to verify that the vehicle is physically occupying a specific parking space while the validation process is being initiated from a location associated with a retail vendor or an alternative evidence source. By requiring this precise temporal and spatial alignment, the system creates a technical handshake between the physical-world transaction and the digital parking record. This multilayered verification effectively mitigates the risk of fraudulent redemptions (such as using evidence sourced from outside a permitted incentive zone or using "stale" evidence from a previous day) thereby providing a secure and automated alternative to manual kiosk-based validation.Attorney Docket No.: EXG-1

[0078] At step 208, the method 200 includes identifying a matched discount offer by comparing the validation qualifying information to a dataset of discount offers each associated with a prescribed parking discount. The comparison may be performed by an analysis engine or rules engine that evaluates the validation qualifying information against criteria set by retail vendors via a vendor management interface. This process may enable a cross-promotional validation hub, mapping evidence from a first commercial sector (e.g., a sporting event) to an incentive in a second commercial sector (e g., a parking facility). The matched discount offer may be subject to dynamic constraints such as budget caps, redemption windows, or specific qualifying events (e.g., a team victory).

[0079] The matching process at step 208 may be executed by a rules engine or matching module configured to evaluate whether the validation qualifying information satisfies a complex hierarchy of threshold conditions, eligibility parameters, and contextual constraints specified within the dataset. This matching logic may operate as an asynchronous service, allowing the system to compare a single piece of arbitrary evidence against a plurality of disparate offers across the cross-promotional network. Importantly, the dataset of discount offers is not static; it is configured to dynamically update in real-time based on market conditions, user engagement metrics, and direct inputs from retail vendors 102 via the vendor management interface. This ensures that the matched discount offer reflects current victory discounts, budget-cap statuses, or time-sensitive promotional windows.

[0080] At step 210, the method 200 includes applying the prescribed parking discount associated with the matched discount offer to a parking fee associated with a parking session. Applying the discount may involve updating the parking fee in a database linked to the parking facility (e.g., parking facility 104). The value of the applied discount may be dynamically adjusted based on a temporal decay7logic, wherein the discount amount reduces as the time elapsed since the qualifying transaction increases. Furthermore, a prioritization algorithm may utilize a historical behavior profile to select or prioritize specific discount offers that maximize user utility7or retail conversion. While described with respect to parking fees, the discount may similarly be applied to other sendee fees, such as EV charging or car wash services.Attorney Docket No.: EXG-1

[0081] At step 212, the method 200 includes causing to initiate a transaction to affect payment of the parking fee. This may occur through a linked parking account storing settlement methods such as credit card information or digital wallet credentials. In some embodiments, payment is affected automatically based on a default setting, allowing the user to complete the entire validation and payment process without physical interaction with a parking payment kiosk (e.g., kiosks 116, 118). Upon successful payment, a release authorization may be generated and transmitted to a local barrier operator to grant egress of the vehicle from the parking facility. Throughout the process, the system may execute a generative model to provide real-time, context-aware notifications or troubleshooting instructions to the user via their mobile device.

[0082] The release authorization generated by the remote server 124 may include an authorization token, a validity7period, and identifying information associated with the vehicle 122, such as a vehicle make, model, color, or license plate number. A local barrier operator having a communication interface receives this release authorization and transmits the data to an onboard control module. The control module is configured to validate the authorization (e g., checking for expiration or revocation) and, upon verification, generate a control signal that is transmitted to an onboard actuator. This actuator initiates a physical release event, such as driving a motor to raise the movable barrier 110 from a closed position to an open position.

[0083] To ensure the physical security of the egress event, the system may execute a computer vision handshake via the camera 120. The camera 120 monitors the area proximate to the movable barrier 110 and compares captured visual data to the identifying information packaged within the release authorization. When the vehicle 122 is confirmed at the barrier, and optionally when a detection module (e.g., an inductive loop or infrared sensor) detects the physical presence of the vehicle, the control signal is triggered. In high-security' or high-value incentive embodiments, the release may be further conditioned upon the user presenting a physical token to a scanner or image capture device at the barrier, providing a layer of multi-factor authentication. Crucially, this technical handshake allows for automated egress without requiring the user to be near the barrier at the time of payment or to interactAttorney Docket No.: EXG-1with a physical payment kiosk 116, 118, thereby streamlining the real-world transition from retail vendor to exit.

[0084] The systems and methods described herein are further characterized by their infrastructure flexibility’. While the method 200 is described in the context of a parking facility' 104 having physical barriers (PARCS), it is equally applicable to ungated parking facilities or facilities operating without physical access control hardware. In such non-PARCS implementations, the release authorization generated at step 212 may manifest as a digital ledger update or a paid status notification transmitted to an enforcement system or a mobile monitoring unit. This allows for end-to-end validation and payment without requiring the driver to interact with any on-site personnel or hardware, effectively providing a "virtual gate" that reduces driver dwell time and enhances the throughput and safety of the facility regardless of its physical configuration.

[0085] FIG. 3 illustrates an example piece of evidence in the form of a receipt 300. The receipt 300 includes several identifying blocks of discernable information, each used by the remote server and the machine-learned evidence model to determine whether a customer qualifies for a discount to be applied to their parking fee or other cross-sector incentive. By way of non-limiting example, the identifying blocks can include a vendor identification block 302, a location block 304, a timestamp block 306, a transaction value block 308, a transaction or receipt barcode 310, or any combination thereof. Yet other types of identifying blocks are possible in various combinations and layouts based on the retail vendor, the particular location, and other factors. As previously described, while the receipt 300 is depicted as a physical transaction document, it may represent any arbitrary evidence, such as a digital screenshot, an electronic boarding pass, or a sports arena ticket, that is inherently devoid of specific indicia identifying a qualification for a parking discount.

[0086] The identification block 302 identifies the retail vendor and is typically arranged at the top of the receipt 300, optionally including the name of the retail vendor, contact information, and a logo. The location block 304 may include a physical address, such as the unit number, street address, city, and state of the retail vendor. The machine-learned evidence model may utilize the identification block 302 and the location block 304 to perform a geospatial correlation, ensuring the retailAttomey Docket No.: EXG-1vendor is a participating member of the cross-promotional network and that the transaction occurred within a permitted geographical proximity to the parking facility’ or another service point.

[0087] The timestamp block 306 can include a time of purchase, i.e., when the transaction was completed and payment was authorized. This timestamp 306 may be cross-referenced with the ingress time of a vehicle to establish a temporal link between the transaction and an active parking session. Furthermore, the system may use the timestamp 306 to calculate a temporal decay for the discount, wherein the value of the prescribed parking discount is dynamically reduced as the time elapsed since the transaction increases.

[0088] The transaction value block 308 may include one or two parts. Two-part value blocks 308 may include an itemized value block 308A and a total value block 308B. The itemized value block 308A includes an individual listing of items or sendees purchased in the transaction. The total value block 308B includes the overall sales price for all of the items listed in the itemized value block 308A, and potentially other fees such as taxes, surcharges, tips and human labor additions, and the like. Some receipts may include only a total value block 308B, omitting the itemized value block 308A. In some embodiments, the machine-learned evidence model identifies qualifying events by analyzing the itemized value block 308A to detect the purchase of specific products or services that trigger high-value incentives, rather than relying solely on the overall sales price in the total value block 308B.

[0089] The transaction or receipt barcode 310 is an internal barcode or indicia used by the retail vendor to identify the transaction, e.g., in the event of a return or dispute, allowing for quick receipt identification using barcode or similar scanning techniques. The barcode 310 is an internal retail identifier and does not contain or reference information intended to provide parking validation. However, the analysis engine may utilize the barcode 310 as a unique transaction identifier to perform fraud detection, such as by checking the barcode 310 against a ledger of previously redeemed evidence to prevent duplicate validation attempts across the network of participating facilities.

[0090] Beyond the specific identifying blocks 302-310, the arbitrary evidence may include various other types of information and / or metadata that the machine-Attorney Docket No.: EXG-1learned evidence model is configured to extract and utilize for validation. For example, in the context of an alternative evidence source such as a sporting event, the information may include event-specific identifiers, such as a game opponent, a seat or section number, a gate entry time, or a unique event serial number. In other implementations, such as a repair order or service invoice, the information may include service-specific identifiers, such as a Vehicle Identification Number (VIN), a license plate number, or a description of services rendered (e.g., "oil change" or "tire rotation"). The machine-learned evidence model may further be configured to identify inventory-level data, such as Stock Keeping Units (SKUs) or Universal Product Codes (UPCs) within the itemized value block 308A, to determine if the user purchased a specific promotional item required for a high-value discount.Additionally, or alternatively, the evidence may contain loyalty or membership metadata, such as a reward member tier or a masked loyalty account number, which the system may use to apply tiered incentives based on the user's standing with the retail vendor. In some embodiments, payment-method indicators, such as the last four digits of a credit card or a digital wallet provider name (e.g., Apple Pay or Google Pay), may be identified to facilitate a cross-reference with the payment information on file for the parking session, providing a seamless zero-trust verification of the transaction's authenticity without requiring the storage of full financial records.

[0091] Referring again to FIG. 2, the captured 202 evidence is transmitted to a remote server, e.g., from a mobile device, by a wired or wireless communication protocol. To transmit the evidence, a mobile client application can establish a network connection via Wi-Fi or cellular and initiate a request to a remote server designated endpoint. Prior to transmission, the image may be compressed, resized, or encrypted to optimize bandwidth and ensure security. Metadata, such as timestamps, geolocation, and user identifiers, may be appended to the payload.

[0092] In some implementations, the evidence can be encrypted prior to transmission. As such, sensitive information is not shared with a client server. In another implementation, based on user selection and preference, certain data transmitted to and / or stored at the server may be treated in one or more ways before it is stored or used, so that personally identifiable information is removed. As an example, a unique and non-personally identify ing identifier can include a randomAttorney Docket No.: EXG-1number, a result of a hash function applied to a user identifier, or some other obfuscated or encrypted identifier which cannot be reverted to obtain the user identifier. In a particular embodiment, the personally identifiable information can be encoded, e.g., using a Base64 encoding technique to convert binary data into ASCII text format by dividing the data into 6-bit groups and representing each group with a printable character. Such encoding enables safe transmission over text-based protocols.

[0093] After, or even during, transmission of the evidence, the mobile client application can generate an update to the user in the form of a notification. In some implementations, the notification is generated using one or more generative models. The generative model(s) can include, for example, a sequence processing model such as a large language model including 10B parameters or more, an autoregressive language model, a machine-learned multi-modal model, or any other machine-learned model configured to provide generative content in response to the validation process and / or a user query. The generative content may provide a text-based update associated with a step of the validation process currently being undertaken at the remote server (e.g., "Analyzing Receipt Data" or "Correlation Successful").

[0094] The method 200 includes identifying a matched discount offer (step 208). In some instances, the remote server may determine that the customer qualifies for multiple discounts. The mobile client application can display these discounts and provide an option for the customer to select between the discounts. Alternatively, the remote server may select a "preferred" or "best" option using context analysis based on predefined selections made by the customer in their settings or based on a learned optimization schema (e.g., selecting the discount with the highest monetary value). Some discount offers may be set by the retail vendor to be combinable subject to certain conditions or metrics.

[0095] In an embodiment, updating 210 the parking fee can include determining a current parking fee based on a parking factor, such as the duration of the current parking session, a vehicle type, a surge upcharge, or a combination thereof. The calculated discount can then be deducted from the current parking fee to arrive at the updated parking fee. Affecting 212 payment can occur through a linked parking account storing a settlement method, including for example, credit card information,Attorney Docket No.: EXG-1bank account details (e.g., digital bank transfer records) for automated clearing house (ACH) transactions, digital wallet credentials (e g., PayPal, Apple Pay, Google Pay), or cryptocurrency wallet addresses, and any tokenized or encrypted representations thereof. Upon successful settlement, the system may generate a release authorization, such as a barcode or other scannable reference, to be scanned at a PARCS system (e.g., a movable barrier operator) to allow the user to exit the parking facility without physical kiosk interaction.

[0096] FIG. 4 illustrates a schematic of a mobile device 400 including a body 402 supporting an image capture device 404, a processor 406 in communication with the image capture device 404 and configured to receive image data therefrom, a mobile client application 408 associated with the validation and cross-promotional functionality, wireless communication circuitry 410, a user input 412, and a display 414. The processor 406 may include a multi-core architecture comprising a central processing unit (CPU) and a graphical processing unit (GPU). In some instances, the processor 406 may include a dedicated neural processing unit (NPU) or Al engine specifically optimized for executing edge-based machine learning inference with low latency and reduced power consumption. The image capture device 404 may include a high-resolution CMOS or CCD sensor coupled with an image signal processor (ISP) configured to perform real-time pixel-level adjustments such as high dynamic range (HDR) processing, noise reduction, and white balance optimization prior to the data reaching the mobile client application 408.

[0097] Referring to FIG. 5, the mobile client application 408 can include an evidence module 500, a display module 502, and a communication module 504. The evidence module 500 is configured as a high-throughput edge-computing engine that performs pre-processing of the image data to identify, e.g., a "Region of Interest" (ROI) containing the arbitrary evidence. In some implementations, the evidence module 500 executes a local lite version of the machine-learned evidence model to perform preliminary character recognition and data integrity checks. The evidence module 500 may further extract a feature vector or a mathematical representation of the evidence rather than a raw pixel map, thereby significantly reducing the payload size for the communication circuitry 410. To safeguard user privacy, the evidence module 500 may utilize a Secure Enclave or Trusted Execution Environment (TEE)Attorney Docket No.: EXG-1within the processor 406 to perform Base64 encoding or SHA-256 hashing of sensitive transaction identifiers, ensuring that no personally identifiable information (Pll) exists in a raw or recoverable state during transit.

[0098] The display module 502 is configured to facilitate a multi-modal humanmachine interface via the display 414. In some embodiments, the display module 502 generates an Augmented Reality (AR) effect that overlays a semi-transparent template on the live viewfinder, guiding the user to align the arbitrary evidence and any required physical token within specific focal zones. The display module 502 may utilize the haptic engine of the body 402 to provide tactile feedback (e.g., a vibration) when the evidence is correctly aligned and focused, serving as a non-visual confirmation that the image is suitable for capture. Furthermore, the display module 502 may be configured to present generative content, such as dynamically rendered UI components (e.g., progress rings or real-time status tickers) that are updated via the communication module 504 as the remote server 124 progresses through the validation steps.

[0099] The communication module 504 manages the link-layer and applicationlayer protocols for data exchange via the wireless communication circuitry 410. The communication module 504 may implement an intelligent transport fallback logic, wherein the system first attempts to transmit the validation data via a secure HTTPS / TLS 1.3 web-based request, but may automatically pivot to an SMS / MMS-based ingest gateway if cellular data bandwidth falls below a threshold or if a native application connection cannot be established. This ensures high system availability7in underground parking structures or areas with degraded network coverage.Additionally, the communication module 504 may be configured to synchronize a living memory cache stored locally on the mobile device 400, allowing the user to view previously validated discounts or stale evidence even when the device is in an offline state.

[0100] In some implementations, the evidence module 500 provides sophisticated surface-integrity analysis to mitigate capture failures. For example, the module may utilize a Laplacian variance method to calculate a blur score for the image, rejecting the capture if the score indicates insufficient sharpness for OCR. Additionally, the module may perform a reflectance and light-direction analysis to identify glare or hotAttorney Docket No.: EXG-1spots caused by overhead garage lighting or camera flash, which might occlude standard identifying blocks (e.g., the timestamp block 306). If such occlusion is detected, the display module 502 may generate a generative multi-modal prompt or a visual arrow directing the user to "Tilt phone to reduce glare." For arbitrary evidence in the form of physical tokens, the evidence module 500 may further analyze microtext or holographic reflectance patterns to verily the authenticity of the token, crossreferencing these visual "fingerprints" with a known library of authorized tokens to prevent the use of digital reproductions or counterfeit artifacts.

[0101] The captured evidence may be handled within a transient data lifecycle designed for maximum security'. Raw image data may be stored in an encrypted, app-specific temporary folder and automatically purged via a secure delete routine (e.g., overwriting with random bits) immediately upon successful transmission or after a short expiration period (e.g., 60 seconds). This minimizes the digital footprint of the transaction on the mobile device 400. In certain embodiments, the mobile client application 408 may utilize the user input 412 to capture a proof of intent signal — such as a biometric scan or a specific gesture — which is appended to the evidence payload as an immutable signature, providing non-repudiation for the payment transaction initiated at step 212 of method 200.

[0102] Referring to FIG. 6, a detailed block diagram illustrates the remote server 124, as seen in accordance with an example embodiment. The remote server 124 can be defined by a hardware boundary comprising one or more processors 128 and memory 130 (as described with respect to FIG. 1), logically and physically partitioned to form a high-throughput, secure data processing pipeline that distinguishes the system as a special-purpose computing machine. This specialized architecture enables the ingestion and semantic analysis of unstructured arbitrary evidence while maintaining strict PII obfuscation at the edge of the processing core.

[0103] As illustrated, the remote server 124 includes an ingest gateway 602 positioned at the network edge to manage data acquisition. The ingest gateway 602 is configured with multiple communication interfaces to accept data representing images from disparate sources via complex, parallel communication protocols. These sources include, but are not limited to, wireless communication circuitry 410 of a mobile device 400 (via a native application or web gateway) and a standard SMS / MMSAttorney Docket No.: EXG-1protocol shortcode or long-code endpoint. The ingest gateway 602 may perform preliminary handshake verification and session identification to link incoming data to an active parking session or user profile.

[0104] Upon acquisition, the raw data may be routed to a Personally Identifiable Information (PII) obfuscation layer 604. This layer 604 functions as a zero-trust security gate, executing advanced cryptographic and encoding routines to anonymize sensitive transaction data before it enters the main processing modules. For example, the PII obfuscation layer 604 may divide binary image data into discrete bit groups and apply a Base64 encoding technique to convert the groups into printable ASCII characters. Alternatively, or additionally, transactional identifiers, such as credit card numbers or serial identifiers, may be hashed via a SHA-256 routine to ensure the data is mathematically irreversible. This obfuscation ensures that sensitive financial or personal data is never stored in a vulnerable, recoverable format within the server 124.

[0105] The secure, obfuscated data payload is transmitted from the PII obfuscation layer 604 to a data processing module 606. The data processing module 606 executes a complex series of normalization and transformation operations to prepare the raw image data for semantic analysis. In a particular embodiment, the data processing module 606 can include a normalization submodule 605 configured to perform optical character recognition (OCR), contrast adjustment, and orientation correction on the image. The module 606 can further, or alternatively, include a Region of Interest (ROI) extraction submodule 607 that identifies the precise geographical zones within the image containing the standard identifying blocks (e.g., the timestamp block 306 or transaction value block 308 of FIG. 3), filtering out visual noise and irrelevant formatting data.

[0106] The normalized, ROI-specific data is transmitted to the analysis engine 608. The analysis engine 608 is configured to perform high-level inference on the unstructured data to derive the validation qualifying information. The analy sis engine 608 comprises a multi-model logic core, including at least a machine-learned (ML) evidence model 610 and a generative multi-modal model 612. The ML evidence model 610 is trained on diverse datasets to recognize the contextual patterns and semantic signatures that distinguish qualifying transactions from non-qualifying onesAttorney Docket No.: EXG-1within the ROI data, even across variable document layouts. The generative multimodal model 612 may include an LLM having 10B parameters or more. This generative model 612 is configured to analyze the context of the validation event and, as described with respect to method 200, produce real-time, natural language notifications, context-aware offers, or interactive troubleshooting prompts for the user.

[0107] A living memory database 614 is in multi-directional communication with the analysis engine 608 via a specialized data bus. The living memory database 614 persists data that is dynamically adjustable and crucial for the real-time operation of the analysis engine 608, including the user's historical behavior profiles and the dynamic discount dataset. The dynamic discount dataset allows the analysis engine 608 to perform real-time rules engine comparisons against current budget caps, redemption windows, and specific qualifying event triggers defined via the vendor management terminal 119. By cross-referencing incoming evidence against the historical behavior profile, the living memory database 614 enables the predictive modeling and relevance scoring logic required for optimized cross-sector promotional mapping.

[0108] Referring to FIG. 7, a data flow diagram illustrates the operation of a validation hub in executing a cross-sector promotional mapping. The validation hub, hosted within the remote server 124. functions as a multi -tenant digital gateway for the distributed, non-Indicia validation ecosystem. In this configuration, the remote server 124 is configured to receive arbitrary evidence 704 from a user, who may capture an image of the evidence 704 while within, near, or leaving a location associated with a first commercial sector 702, which may comprise a business or event venue unrelated to the final validation point, such as a sports stadium or arena.

[0109] The arbitrary evidence 704 is visualized as a conceptual commercial artifact, such as a ticket stub or e-receipt, which is devoid of specific indicia identifying a qualification for a parking discount. The remote server 124 (via the machine-learned evidence model 610, for example) performs semantic analysis on the input data representing the arbitrary evidence 704 to identify a qualifying event (e g., a commercial transaction exceeding a threshold or attendance at a specific event like a "victory' game").Attorney Docket No.: EXG-1

[0110] Upon determining qualification, the validation hub maps 706 this qualifying event across commercial sectors to identify an appropriate downstream incentive 710. As visualized by arrow 710, this incentive may comprise a dynamic parking fee update or a digital token generation for use at a second commercial sector 708, which may comprise a neighboring retail vendor 102 or the parking facility 104 itself. This automated mapping allows the remote server 124 to dynamically update a parking session of the user linked to the evidence without requiring manual point-of-sale interaction by the retail vendor 102 or physical voucher redemption at a parking kiosk. The cross-sector data flow may include redemption data and / or a budget update path visualized as a loopback arrow from a second commercial sector 708 back to the validation hub 706. This feedback loop allows the validation hub 706 to dynamically track the cumulative value of redeemed incentives against budget caps or redemption window s defined by the retail vendor via the vendor management interface, thereby automatically suspending or adjusting future discount offers to manage vendor liability and facility’ throughput.

[0111] Further aspects of the invention are provided by one or more of the following embodiments:

[0112] A non-transitory computer-readable medium storing instructions which, when executed by one or more processors, causes performance of operations, the operations comprising: receiving, at a remote server, data representing an image of evidence associated with a transaction between a user and a vendor, wherein the evidence is devoid of indicia specifically identifying a qualification for a parking discount; analyzing, at the remote server, the data to identify' discernable information within the evidence, the discernable information including at least one of a vendor identifier, a transaction location, a timestamp, or a transaction value; generating validation qualifying information by processing the discernable information through a machine-learned evidence model; matching the validation qualify ing information against a dataset of discount offers to identify a prescribed parking discount; and causing to update a parking fee associated with a parking session of the user to include the prescribed parking discount.

[0113] The computer-readable medium of any one or more of the embodiments, wherein the evidence comprises a physical receipt of purchase, a digital bank transferAttorney Docket No.: EXG-1record, an invoice, an order confirmation, an email notification, or a store loyalty program record.

[0114] The computer-readable medium of any one or more of the embodiments, wherein the operations further comprise performing, at a mobile device associated with the user, at least one of: document edge detection, perspective correction, wrinkle and crease detection, and shadow analysis to validate the quality of the image prior to transmission of the data to the remote server.

[0115] The computer-readable medium of any one or more of the embodiments, wherein the discernable information is identified within one or more discrete blocks of the evidence, the blocks comprising a vendor identification block, a location block, a timestamp block, or a transaction value block.

[0116] The computer-readable medium of any one or more of the embodiments, wherein the machine-learned evidence model is trained on a dataset comprising historical data samples and ground truth labels to identify patterns and correlations within unstructured transaction data.

[0117] The computer-readable medium of any one or more of the embodiments, wherein the operations further comprise executing a large language model (LTM) to generate a context-aware notification for the user based on at least one of: the status of the validation process, a description of the identified discount, or troubleshooting instructions.

[0118] The computer-readable medium of any one or more of the embodiments, wherein updating the parking fee further comprises automatically initiating an account settlement transaction via a linked parking account storing at least one of: credit card information, bank account details for ACH transactions, or digital wallet credentials.

[0119] The computer-readable medium of any one or more of the embodiments, wherein the operations further comprise generating a release authorization including vehicle identifying information, the vehicle identifying information comprising at least one of: a license plate number, a vehicle make, a vehicle model, or a vehicle color.

[0120] The computer-readable medium of any one or more of the embodiments, wherein the dataset of discount offers includes a plurality of data subsets, each dataAttorney Docket No.: EXG-1subset associated with a different retail vendor and independently adjustable by the associated retail vendor.

[0121] A parking validation system comprising: a communication interface configured to receive data representing an image of evidence associated with a transaction between a user and a retail vendor, wherein the evidence is devoid of indicia specifically identifying a qualification for a parking discount; a memory' storing a model and a dataset of discount offers; and one or more processors in communication with the communication interface and the memory, the one or more processors configured to: identify discernable information within the data, the discernable information including at least one of a vendor identifier, a transaction location, a timestamp, or a transaction value; execute the model to generate validation qualifying information based on the discernable information; compare the validation qualifying information to the dataset of discount offers to identify' a matched discount offer; and update a parking fee associated with a parking session of the user based on a prescribed parking discount associated with the matched discount offer.

[0122] The system of any one or more of the embodiments, further comprising a vendor management interface.

[0123] The system of any one or more of the embodiments, wherein the vendor management interface is configured to receive a predefined budget cap from the retail vendor, and wherein the one or more processors are configured to automatically suspend the matched discount offer in the dataset when a total value of applied discounts reaches the predefined budget cap.

[0124] The system of any one or more of the embodiments, wherein the system is configured to facilitate the update of the parking fee and the payment of the discounted parking fee without physical interaction between the user and a parking payment kiosk.

[0125] The system of any one or more of the embodiments, wherein the one or more processors are configured to facilitate the update of the parking fee and the payment of the discounted parking fee while a mobile device that transmitted the data is at a location remote from the parking facility.

[0126] The system of any one or more of the embodiments, wherein the one or more processors are further configured to: perform a multi-modal temporal-geospatialAttorney Docket No.: EXG-1correlation between the timestamp and transaction location identified within the evidence and a start time and a physical location of the parking session; and validate the matched discount offer only upon verifying, via a real-time signal from a parking spot identification system, that a vehicle associated with the parking session is currently occupying a specific parking space within the parking facility7.

[0127] A method for validating a discount for a parking fee, the method comprising: receiving, at a remote server, data representing an image of evidence associated with a transaction between a user and a retail vendor, the evidence being devoid of specific indicia identifying a qualification for a parking discount; identifying, via one or more processors at the remote server, discernable information within the data, the discernable information including at least one of a vendor name, a transaction location, a timestamp, or a transaction value; determining validation qualifying information by processing the discernable information through a machine-learned evidence model; identifying a matched discount offer by comparing the validation qualifying information to a dataset of discount offers each associated with a prescribed parking discount; applying the prescribed parking discount to a parking fee associated with a parking session; and causing to initiate a transaction to affect payment of the parking fee.

[0128] The method of claim 16, further comprising: executing an LLM to generate a natural language notification for the user based on at least one of the discernable information, the matched discount offer, or the prescribed parking discount; and transmitting the natural language notification to a mobile device associated with the user.

[0129] The method of any one or more of the embodiments, further comprising encoding personally identifiable information within the data using a Base64 encoding technique to convert binary data into an ASCII text format prior to identifying the discernable information.

[0130] The method of any one or more of the embodiments, wherein applying the prescribed parking discount further includes calculating the parking fee based on a surge upcharge or a duration of the parking session.Attorney Docket No.: EXG-1

[0131] The method of any one or more of the embodiments, further comprising discarding the data representing the image and the discernable information after the validation qualifying information is determined.

[0132] A non-transitory computer-readable medium storing instructions which, when executed, causes performance of a method of validating a discount for a parking fee, the method comprising: receiving, at a remote server, data from an image of a receipt taken by a camera of a mobile device; analyzing, at the remote server, the data to recognize discernable information; analyzing, at the remote server, the discernable information using a machine-learned evidence model configured to generate validation qualifying information; comparing, at the remote server, the validation qualifying information to a dataset including one or more discount offers each associated with a prescribed parking discount; applying, at the remote server, the prescribed discount to a parking fee associated with a parking session based on the comparing; affecting payment for the discounted parking fee; and generating a release authorization after payment is affected, the release authorization granting egress of a vehicle associated with the parking session from a parking facility.

[0133] The non-transitory computer-readable medium of any one or more of the embodiments, further comprising discarding the image and the discernable information after analyzing is complete.

[0134] The non-transitory computer-readable medium of any one or more of the embodiments, wherein affecting payment comprises: transmitting, from the remote server, a payment authorization request to a payment processing system, the payment authorization request associated with the discounted parking fee; receiving, at the remote server, a response from the payment processing system indicating approval or rejection of the payment transaction; and updating, at the remote server, a transaction status based on the response from the payment processing system, wherein generating the release authorization occurs in response to updating the transaction status.

[0135] The non-transitory computer-readable medium of any one or more of the embodiments, wherein generating the release authorization comprises causing a notification to be generated on the mobile device, the notification confirming the release authorization.Attorney Docket No.: EXG-1

[0136] The non-transitory computer-readable medium of any one or more of the embodiments, further comprising executing a large language model (LLM) to generate the notification based on the discernable information, the validation qualifying information, the discount offers, the prescribed discount, the received payment, the release authorization, or any combination thereof.

[0137] The non-transitory computer-readable medium of any one or more of the embodiments, wherein generating the release authorization further comprises transmitting information associated with the release authorization to a transceiver associated with the parking facility, and wherein the parking facility is configured to actuate a barrier operator to release the vehicle from the parking facility based on the received information associated with the release authorization.

[0138] The non-transitory computer-readable medium of any one or more of the embodiments, wherein the dataset comprises a stored list of discount offers and associated qualify ing events, wherein comparing the validation qualify ing information to the dataset includes comparing the validation qualifying information to the associated qualifying events and selecting an appropriate discount offer from the stored list of discount offers based on the comparing.

[0139] The non-transitory computer-readable medium of any one or more of the embodiments, wherein the stored list of discount offers and associated qualify ing events includes one or more data subsets, each data subset associated with a different retail vendor, and wherein each of the data subsets is adjustable by the associated retail vendor.

[0140] The non-transitor7computer-readable medium of any one or more of the embodiments, wherein the receipt excludes indicia associated with the validation qualifying information or the one or more discount offers.

[0141] A method of validating a discount for a parking fee, the method comprising: capturing, using a mobile client application executed by a mobile device, an image of a receipt from a retail vendor; transmitting, from the mobile device, the captured image to a remote server; receiving, at the mobile device, a signal from the remote server, wherein the signal includes information that a current parking session qualifies for a discount based on the captured image; updating a parking fee associated with the parking session to include the discount; and generating, on aAttorney Docket No.: EXG-1display of the mobile device, a notification indicative of the discounted parking fee; receiving, at the mobile device, a user input accepting the discounted parking fee; and affecting payment of the discounted parking fee in response to receiving the user input.

[0142] The method of any one or more of the embodiments, wherein affecting payment of the discounted parking fee is performed without any physical interaction with parking facility equipment.

[0143] The method of any one or more of the embodiments, wherein generating the notification indicative of the discounted parking fee is executed by a large language model (LLM) in view of information extracted from the image, the discount, an associated qualifying event, or any combination thereof.

[0144] The method of any one or more of the embodiments, wherein the receipt excludes indicia associated with the parking session and the discount.

[0145] This written description uses examples to disclose the invention, including the best mode, and also to enable any person skilled in the art to practice the invention, including making and using any devices or systems and performing any incorporated methods. The patentable scope of the invention is defined by the claims, and may include other examples that occur to those skilled in the art. Such other examples are intended to be within the scope of the claims if they include structural elements that do not differ from the literal language of the claims, or if they include equivalent structural elements with insubstantial differences from the literal language of the claims.

Claims

Attorney Docket No.: EXG-1WHAT IS CLAIMED IS:

1. A non-transitory computer-readable medium storing instructions which, when executed by one or more processors, causes performance of operations, the operations comprising:receiving, at a remote server, data representing an image of evidence associated with a transaction between a user and a vendor, wherein the evidence is devoid of indicia specifically identifying a qualification for a parking discount;analyzing, at the remote server, the data to identify discernable information within the evidence, the discernable information including at least one of a vendor identifier, a transaction location, a timestamp, or a transaction value;generating validation qualifying information by processing the discernable information through a machine-learned evidence model; matching the validation qualifying information against a dataset of discount offers to identify a prescribed parking discount; andcausing to update a parking fee associated with a parking session of the user to include the prescribed parking discount.

2. The computer-readable medium of claim 1, wherein the evidence comprises a physical receipt of purchase, a digital bank transfer record, an invoice, an order confirmation, an email notification, or a store loyalty program record.

3. The computer-readable medium of claim 1, wherein the operations further comprise performing, at a mobile device associated with the user, at least one of: document edge detection, perspective correction, wrinkle and crease detection, and shadow analysis to validate the quality of the image prior to transmission of the data to the remote server.

4. The computer-readable medium of claim 1, wherein the discernable information is identified within one or more discrete blocks of the evidence, theAttorney Docket No.: EXG-1blocks comprising a vendor identification block, a location block, a timestamp block, or a transaction value block.

5. The computer-readable medium of claim 1, wherein the machine-learned evidence model is trained on a dataset comprising historical data samples and ground truth labels to identify patterns and correlations within unstructured transaction data.

6. The computer-readable medium of claim 1, wherein the operations further comprise executing a large language model (LLM) to generate a context-aware notification for the user based on at least one of: a status of the validation process, a description of the prescribed parking discount, or troubleshooting instructions.

7. The computer-readable medium of claim 1, wherein updating the parking fee further comprises automatically initiating an account settlement transaction via a linked parking account storing at least one of: credit card information, bank account details for ACH transactions, or digital wallet credentials.

8. The computer-readable medium of claim 1 , wherein the operations further comprise generating a release authorization including vehicle identifying information, the vehicle identifying information comprising at least one of: a license plate number, a vehicle make, a vehicle model, or a vehicle color.

9. The computer-readable medium of claim 1, wherein the dataset of discount offers includes a plurality of data subsets, each data subset associated with a different retail vendor and independently adjustable by the associated retail vendor.

10. A parking validation system comprising:a communication interface configured to receive data representing an image of evidence associated with a transaction between a user and a retail vendor, wherein the evidence is devoid of indicia specifically identifying a qualification for a parking discount;Attorney Docket No.: EXG-1a memory' storing a model and a dataset of discount offers; andone or more processors in communication with the communication interface and the memory', the one or more processors configured to: identify discernable information within the data, the discernable information including at least one of a vendor identifier, a transaction location, a timestamp, or a transaction value; execute the model to generate validation qualifying information based on the discernable information;compare the validation qualifying information to the dataset of discount offers to identify a matched discount offer; and cause to update a parking fee associated with a parking session of the user based on a prescribed parking discount associated with the matched discount offer.

11. The system of claim 10, further comprising a vendor management interface.

12. The system of claim 11, wherein the vendor management interface is configured to receive a predefined budget cap from the retail vendor, and wherein the one or more processors are configured to automatically suspend the matched discount offer in the dataset when a total value of applied discounts reaches the predefined budget cap.

13. The system of claim 10, wherein the system is configured to update of the parking fee without physical interaction between the user and a parking pay ment kiosk.

14. The system of claim 13, wherein the one or more processors are configured to facilitate the update of the parking fee and the payment of the discounted parking fee while a mobile device that transmitted the data is at a location remote from a parking facility.Attorney Docket No.: EXG-115. The system of claim 10, wherein the one or more processors are further configured to:perform a multi-modal temporal-geospatial correlation between the timestamp and transaction location identified within the evidence and a start time a physical location of the parking session; andvalidate the matched discount offer only upon verifying, via a real-time signal from a parking spot identification system, that a vehicle associated with the parking session is currently occupying a specific parking space within a parking facility.

16. A method for validating a discount for a parking fee, the method comprising:receiving, at a remote server, data representing an image of evidence associated with a transaction between a user and a retail vendor, the evidence being devoid of specific indicia identifying a qualification for a parking discount;identifying, via one or more processors at the remote server, discernable information within the data, the discernable information including at least one of a vendor name, a transaction location, a timestamp, or a transaction value;determining validation qualifying information by processing the discernable information through a machine-learned evidence model; identifying a matched discount offer by comparing the validation qualifying information to a dataset of discount offers each associated with a prescribed parking discount;applying the prescribed parking discount to a parking fee associated with a parking session; andcausing to initiate a transaction to affect payment of the parking fee.

17. The method of claim 16, further comprising:Attorney Docket No.: EXG-1executing an LLM to generate a natural language notification for the user based on at least one of the discernable information, the matched discount offer, or the prescribed parking discount; and transmitting the natural language notification to a mobile device associated with the user.

18. The method of claim 16, further comprising encoding personally identifiable information within the data using a Base64 encoding technique to convert binary data into an ASCII text format prior to identifying the discernable information.

19. The method of claim 16, wherein applying the prescribed parking discount further includes calculating the parking fee based on a surge upcharge or a duration of the parking session.

20. The method of claim 16, further comprising discarding the data representing the image and the discernable information after the validation qualifying information is determined.

Images