KR-20260068128-A - AUTOMATED ARTIFICIAL INTELLIGENCE VEHICLE APPRAISALS

Abstract

A vehicle appraisal system that uses a dynamic interface equipped with vehicle capture modules to capture image and audio data of a vehicle and processes the image and audio data to automatically calculate vehicle metrics. The system uses the vehicle metrics to generate cost data and market value estimates for the vehicle. The system can integrate with other systems using an application programming interface to exchange data and reports.

Inventors

• 사우딘 스티븐 알.

Assignees

• 디스커버리 로프트 인코포레이티드

Dates

Publication Date

20260513

Application Date

20191016

Priority Date

20181019

Claims (20)

1. In a system for vehicle identification using image processing, A server having a non-transient computer-readable storage medium, wherein the non-transient computer-readable storage medium comprises one or more processors: An interface application having a vehicle capture module that captures images of a vehicle and captures metadata for said captured images, wherein the interface application displays an interactive guide to help capture said images, said interactive guide has overlays that update to help capture different images of said views of the vehicle, said interactive guide is generated using a cage for a vehicle type, said cage defines locations or components of said vehicle, said component of said vehicle includes at least one tire, said vehicle identification number is metadata for said captured images, said vehicle identification number indicates said vehicle type, said overlays include different cage views to help capture at least some of the different images of said views of the vehicle; At least one agent interface, wherein each agent interface has a task dashboard for displaying a portion of the captured images for receiving input data, and the at least one agent interface displays the different cage views as overlays on at least a portion of the different images of the vehicle views; A recognition engine that processes the captured images and metadata to detect defects of the vehicle and calculate vehicle metrics, wherein the vehicle metrics include tire data, the processing is based on at least one task sent to the at least one agent interface to receive input data for detecting defects of the vehicle and calculating the vehicle metrics, each task is sent to the corresponding agent interface, each task is associated with the portion of the captured images to be displayed within the corresponding agent interface, the system defines tasks for each view of the vehicle, and the recognition engine calculates the tire data by processing the captured images using the cage to identify the tire among the captured images and to link the tire data to at least one individual tire of the vehicle, wherein the tire data includes the tread depth of the at least one individual tire; A cost estimation tool that processes the vehicle metrics to calculate cost data for repairing defects of the vehicle; and A valuation tool that calculates a market value estimate for the vehicle using the above vehicle metrics and the above cost data; A system comprising the server having executable instructions that configure the
2. In paragraph 1, The above recognition engine detects the remaining tread life of at least one individual tire based on the tread depth of at least one individual tire, a system.
3. In paragraph 1, The above recognition engine is a system that calculates the operability of the vehicle based on the detected defects of the vehicle and the calculated vehicle metrics.
4. In paragraph 1, The above tire data includes a determined damage type of at least one individual tire, in a system.
5. In paragraph 1, The above recognition engine is a system that classifies the tread depth of at least one individual tire into one of a like-new tread, a showing wear tread, and a low tread.
6. In paragraph 1, The above recognition engine is a system that detects at least one of signs of weathering and a tire type.
7. In paragraph 1, A system in which the interface application is configured to receive an error message from the vehicle capture module when it is determined that the tire data cannot be successfully calculated from the captured images, and in response to this, updates the interface application with the visual elements corresponding to the interactive guide to recapture the images for calculating the tire data.
8. In paragraph 1, A system that, when the above tire data cannot be automatically calculated, sends a task to an agent along with at least some of the captured images to receive input data for the above tire, wherein some of the captured images correspond to at least one individual tire.
9. In paragraph 1, The recognition engine calculates the vehicle identification number by decoding the vehicle identification number from the captured images to perform verification, and if the vehicle identification number cannot be automatically decoded, sends a task to an agent along with at least some of the captured images to receive input data for the vehicle identification number as a response, wherein some of the captured images correspond to a vehicle identification number plate.
10. In paragraph 1, The above interface application is a system that dynamically configures the vehicle capture module based on the vehicle type to generate the interactive guide corresponding to the cage.
11. In a method for automatically processing vehicle images, Step of receiving a vehicle identification number from an interface application; A step of displaying an interactive guide in the interface application to capture an image of a vehicle - the interactive guide is created using a cage for a vehicle type, the cage defines the locations or components of the vehicle, at least one component includes at least one tire, a vehicle identification number indicates the vehicle type, the interactive guide has overlays that update to help capture different images of views of the vehicle, the cage has different cage views, and the overlays include different cage views to help capture at least some of the different images of views of the vehicle - ; A step of capturing images of the vehicle and metadata for the captured images in the above interface application - the captured images identify defects in the vehicle - ; A step of displaying a portion of the captured images in at least one agent interface to receive input data - the at least one agent interface displays the different cage views as overlays on at least a portion of the different images of the vehicle views - ; A step of processing the captured images to automatically detect defects of the vehicle and calculate the vehicle metrics—the vehicle metrics include tire data, the processing is to send different tasks to the at least one agent interface and, in response, receive input data for detecting vehicle defects and calculating vehicle metrics, each task is sent to the corresponding agent interface and each task is associated with the portion of the captured images to be displayed within the corresponding agent interface, the system defines tasks for each view of the vehicle, the tire data is calculated by processing the captured images using the cage to identify the tire among the captured images and to associate the tire data with at least one individual tire of the vehicle, the tire data includes the tread depth of the at least one individual tire—; A step of calculating cost data for repairing defects of the above vehicle; and A step of calculating a market value estimate for the vehicle using the above vehicle metrics and the above cost data; A method including
12. In Paragraph 11, A method further comprising the step of detecting the remaining tread life of at least one individual tire based on the tread depth of at least one individual tire.
13. In Paragraph 11, A method further comprising the step of calculating the mobility of the vehicle based on the detected defects of the vehicle and the calculated vehicle metrics.
14. In Paragraph 11, The above tire data includes a determined type of damage of at least one individual tire, a method.
15. In Paragraph 11, A method further comprising the step of classifying the tread depth of at least one individual tire into one of new level tread, wear progress tread, and low tread.
16. In Paragraph 11, A method further comprising the step of detecting at least one of signs of weathering and tire type.
17. In Paragraph 11, A method wherein the interface application is configured to receive an error message from the vehicle capture module when it is determined that the tire data cannot be successfully calculated from the captured images, and in response thereto, the interface application is updated with the visual elements corresponding to the interactive guide to recapture the images for calculating the tire data.
18. In Paragraph 11, A method in which, when the above tire data cannot be automatically calculated, a task is sent to an agent along with at least some of the captured images to receive input data for the above tire, and the parts of the captured images correspond to at least one individual tire.
19. In Paragraph 11, The method further comprises: a step of calculating the vehicle identification number by decoding the vehicle identification number from the captured images; a step of verifying the decoded vehicle identification number; and, if the vehicle identification number cannot be automatically decoded, a step of sending a task to an agent along with at least some of the captured images to receive input data for the vehicle identification number as a response. A method in which some of the above-mentioned captured images correspond to vehicle identification number plates.
20. In Paragraph 11, A method in which the above interface application dynamically configures the vehicle capture module based on the vehicle type to generate the interactive guide corresponding to the cage.

Description

Automated Artificial Intelligence Vehicle Appraisals The present disclosure generally relates to computing platforms, artificial intelligence, computer vision, and image and audio processing. The embodiments described herein relate to systems and processes for evaluating a vehicle for the purpose of verifying its condition and/or evaluating external or mechanical defects or other visually and audibly noticeable repairs requiring attention while recalibrating the vehicle to calculate the accurate market value of the vehicle using computer vision and audio detection. Figure 1 is a diagram of a system for vehicle inspection. FIG. 2 is an exemplary interface of a system for vehicle appraisal according to some embodiments. FIG. 3 is an exemplary interface of a system for vehicle appraisal according to some embodiments. FIG. 4 is an exemplary interface of a system for vehicle appraisal according to some embodiments. FIG. 5 is an exemplary interface of a system for vehicle appraisal according to some embodiments. FIG. 6 is an exemplary interface of a system for vehicle appraisal according to some embodiments. FIG. 7 is an exemplary interface of a system for vehicle appraisal according to some embodiments. FIG. 8 is an exemplary interface of a system for vehicle appraisal according to some embodiments. FIG. 9 is an exemplary interface of a system for vehicle appraisal according to some embodiments. FIG. 10 is an exemplary interface of a system for vehicle appraisal according to some embodiments. FIG. 11 is an exemplary interface of a system for vehicle assessment according to some embodiments. FIG. 12 is an exemplary interface of a system for vehicle assessment according to some embodiments. FIG. 13 is an exemplary interface of a system for vehicle appraisal according to some embodiments. FIG. 14 is an exemplary interface of a system for vehicle assessment according to some embodiments. FIG. 15 is an exemplary interface of a system for vehicle assessment according to some embodiments. FIG. 16 is an exemplary interface of a system for vehicle appraisal according to some embodiments. FIG. 17 is an exemplary interface of a system for vehicle assessment according to some embodiments. FIG. 18 is an exemplary interface of a system for vehicle appraisal according to some embodiments. FIG. 19 is an exemplary interface of a system for vehicle assessment according to some embodiments. FIG. 20 is an exemplary interface of a system for vehicle appraisal according to some embodiments. FIG. 21 is an exemplary interface of a system for vehicle assessment according to some embodiments. FIG. 22 is an exemplary interface of a system for vehicle assessment according to some embodiments. FIG. 23 is an exemplary interface of a system for vehicle appraisal according to some embodiments. FIG. 24 is an exemplary interface of a system for vehicle assessment according to some embodiments. FIG. 25 is an exemplary interface of a system for vehicle appraisal according to some embodiments. FIG. 26 is an exemplary interface of a system for vehicle appraisal according to some embodiments. FIG. 27 is an exemplary interface of a system for vehicle appraisal according to some embodiments. FIG. 28 is an exemplary interface of a system for vehicle assessment according to some embodiments. FIG. 29 is an exemplary interface of a system for vehicle appraisal according to some embodiments. FIG. 30 is an exemplary interface of a system for vehicle appraisal according to some embodiments. Figure 31 is an exemplary process for calculating vehicle metrics from captured images. FIG. 32 is an exemplary process for calculating vehicle metrics from captured images. FIG. 33 is an exemplary process for calculating vehicle metrics from captured images. Figure 34 is an exemplary process for calculating vehicle metrics from captured images. FIG. 35 is an exemplary process for calculating vehicle metrics from captured images. Figure 36 is an exemplary process for calculating vehicle metrics from captured images. Figure 37 is an exemplary image of a vehicle identification number plate. Fig. 38 is an exemplary task (or agent) dashboard for internal data. FIG. 39 is an exemplary task (or agent) dashboard for instrument data or driving data. FIG. 40 is an exemplary interface having input fields for different view data metrics. Figure 41 is an exemplary task (or agent) dashboard for view data. FIG. 42 is an exemplary interface having input fields for front view data metrics. Fig. 43 is an exemplary task (or agent) dashboard for tire data. Figure 44 is an exemplary task (or agent) dashboard for windshield data. Figures 45 and 46 are exemplary interfaces for a task (or agent) dashboard. FIG. 47 is an exemplary interface for a task (or agent) dashboard that may include a list of detected damages. FIG. 48 is an exemplary interface having an interactive guide including a cage overlay for capturing data. FIG. 49 is an exemplary interface (4900) for a cage storage manager. FIG. 50