Vehicle detection and identification by audio propagation

Abstract

A method for motor vehicle detection and identification through audio propagation, comprising: generating a source autometric signature; broadcasting the autometric signature; detecting the autometric signature; visualizing the autometric signature; determining a known autometric signature; and evaluating and sending the autometric signature to a motor vehicle.

Description

Technical Field

[0001] This disclosure relates to vehicle detection and identification. More specifically, this disclosure relates to vehicle detection and identification via audio transmission. Background Technology

[0002] As motor vehicles become more automated and the interfaces between users and vehicles increase, vehicles are increasingly expanding their roles beyond passenger transport. For example, auditory signaling mechanisms emitted by vehicles have the ability to send information during emergencies and are enabling interactive real-world gaming applications.

[0003] While current vehicle recognition systems have achieved their intended purpose, a new and improved system and method for vehicle detection and recognition is needed. Summary of the Invention

[0004] According to several aspects, a method for detecting and identifying motor vehicles via audio transmission includes generating a source automatic measurement signature; broadcasting the automatic measurement signature; detecting the automatic measurement signature; visualizing the automatic measurement signature; identifying a known automatic measurement signature; and evaluating and sending the automatic measurement signature to the motor vehicle.

[0005] In another aspect of this disclosure, the method also includes determining the source of the automatically measured signature data.

[0006] In another aspect of this disclosure, motor vehicles communicate with information exchange infrastructure.

[0007] In another aspect of this disclosure, the method also includes using deep learning image comparison to detect image similarity of signatures.

[0008] In another aspect of this disclosure, the method also includes listening to an automatic measurement signal using a mobile device.

[0009] In another aspect of this disclosure, the mobile device communicates with the information exchange infrastructure.

[0010] In another aspect of this disclosure, generating an automatic metric signature includes a character-to-frequency mapping.

[0011] In another aspect of this disclosure, generating an automatic measurement signature includes a character-to-time mapping.

[0012] In another aspect of this disclosure, generating an automatic measurement signature includes a character-to-color mapping.

[0013] In another aspect of this disclosure, visualized automatic signature measurement includes audio signature analysis and signal data visualization.

[0014] In another aspect of this disclosure, the method also includes storing data associated with the automatic measurement signature in a source repository.

[0015] According to several aspects, the method for detecting and identifying motor vehicles via audio transmission includes generating source automatic measurement signatures, including character-to-frequency mapping, character-to-time mapping, and character-to-color mapping; broadcasting the automatic measurement signatures and detecting the automatic measurement signatures using devices; visualizing the automatic measurement signatures, including audio signature analysis and signal data visualization; identifying known automatic measurement signatures; and evaluating and sending the automatic measurement signatures to motor vehicles.

[0016] In another aspect of this disclosure, the method also includes determining the source of the automatically measured signature data.

[0017] In another aspect of this disclosure, motor vehicles communicate with information exchange infrastructure.

[0018] In another aspect of this disclosure, the method also includes using deep learning image comparison to detect image similarity of signatures.

[0019] In another aspect of this disclosure, the method also includes listening to an automatically measured signature using a mobile device.

[0020] In another aspect of this disclosure, the mobile device communicates with the information exchange infrastructure.

[0021] In another aspect of this disclosure, the method also includes storing data associated with the automatic measurement signature in a source repository.

[0022] According to several aspects, a method for detecting and identifying motor vehicles via audio transmission includes generating a source automatic measurement signature, including character-to-frequency mapping, character-to-time mapping, and character-to-color mapping; visualizing the automatic measurement signature, including audio signature analysis and signal data visualization; identifying known automatic measurement signatures; detecting image similarity of the signatures using deep learning image comparison; and sending the automatic measurement signature to a motor vehicle. The motor vehicle communicates with an information exchange infrastructure, and a mobile device listens to the automatic measurement signal and communicates with the information exchange infrastructure.

[0023] In another aspect of this disclosure, the method also includes evaluating sequences of chords to determine distinct audio signatures.

[0024] The present invention also includes the following solutions:

[0025] Solution 1. A method for detecting and identifying motor vehicles via audio transmission, the method comprising:

[0026] Generate source auto-measure signature;

[0027] Broadcast automatic signature measurement;

[0028] Detect the automatic measurement signature;

[0029] Visualize the automatically measured signature;

[0030] Determine the known automatic measurement signature; and

[0031] The system evaluates and sends an automatic measurement signature to the motor vehicle.

[0032] Option 2. The method described in Option 1 further includes determining the source of the automatically measured signature data.

[0033] Option 3. The method according to Option 1, wherein the motor vehicle communicates with the information exchange infrastructure.

[0034] Option 4. The method according to Option 1 further includes using deep learning image comparison to detect image similarity of the signature.

[0035] Option 5. The method according to Option 1 further includes using a mobile device to listen to an automatic measurement signal.

[0036] Option 6. The method according to Option 5, wherein the mobile device communicates with the information exchange infrastructure.

[0037] Option 7. The method according to Option 1, wherein generating an automatic measurement signature includes a character-to-frequency mapping.

[0038] Option 8. The method described in Option 1, wherein generating an automatic measurement signature includes a character-to-time mapping.

[0039] Option 9. The method according to Option 1, wherein generating an automatic measurement signature includes a character-to-color mapping.

[0040] Option 10. According to the method of Option 1, wherein visualizing the automatic measurement signature includes audio signature analysis and signal data visualization.

[0041] Option 11. The method according to Option 1 further includes storing the data associated with the automatic measurement signature in a source repository.

[0042] Option 12. A method for detecting and identifying motor vehicles via audio transmission, the method comprising:

[0043] The source is automatically measured signature, including character-to-frequency mapping, character-to-time mapping, and character-to-color mapping;

[0044] Broadcast the automatic measurement signature and use the device to detect the automatic measurement signature;

[0045] Visualize the automated signature measurement, including audio signature analysis and signal data visualization;

[0046] Determine the known automatic measurement signature; and

[0047] The system evaluates and sends an automatic measurement signature to the motor vehicle.

[0048] Option 13. The method according to Option 12 further includes determining the source of the automatically measured signature data.

[0049] Option 14. The method according to Option 12, wherein the motor vehicle communicates with an information exchange infrastructure.

[0050] Option 15. The method according to Option 14 further includes using deep learning image comparison to detect image similarity of the signature.

[0051] Option 16. The method according to Option 14 further includes using a mobile device to listen to an automatic measurement signal.

[0052] Option 17. The method according to Option 16, wherein the mobile device communicates with the information exchange infrastructure.

[0053] Option 18. The method according to Option 12 further includes storing the data associated with the automatic measurement signature in a source repository.

[0054] Solution 19. A method for detecting and identifying motor vehicles via audio transmission, the method comprising:

[0055] The source is automatically measured signature, including character-to-frequency mapping, character-to-time mapping, and character-to-color mapping;

[0056] Visualize the automated signature measurement, including audio signature analysis and signal data visualization;

[0057] Visualize the automatically measured signature;

[0058] Identify known automatic measurement signatures;

[0059] Utilizing deep learning image comparison to detect the image similarity of the signature; and

[0060] The automatic measurement signature is sent to the motor vehicle.

[0061] The motor vehicle communicates with the information exchange infrastructure, and the mobile device listens to the automatic measurement signal and communicates with the information exchange infrastructure.

[0062] Option 20. The method according to Option 19 further includes evaluating the sequence of chords to determine different audio signatures.

[0063] Further areas of application will become apparent from the description provided herein. It should be understood that the descriptions and specific examples are intended for illustrative purposes only and are not intended to limit the scope of this disclosure. Attached Figure Description

[0064] The accompanying drawings described herein are for illustrative purposes only and are not intended to limit the scope of this disclosure in any way.

[0065] Figure 1 This is a block diagram of a system for vehicle detection and identification according to an exemplary embodiment;

[0066] Figure 2 This is an application based on exemplary embodiments. Figure 1 An image of the automated measurement feasibility test of the system shown;

[0067] Figure 3 This is an application based on exemplary embodiments. Figure 1 The system shown in the diagram represents the data encoding frequency audio pitch of the audio frequencies; and

[0068] Figure 4 and Figure 5 This is an application based on exemplary embodiments. Figure 1 The diagram shows the data encoding rhythmic audio pitch generated by the system shown. Detailed Implementation

[0069] The following description is exemplary in nature and is not intended to limit this disclosure, application, or use.

[0070] refer to Figure 1 The diagram illustrates system 10, which enables vehicle 12 to send a uniquely identifiable audio signal 16 as a game enabler or to transmit data when vehicle 12 is not easily accessible. Detection devices, such as mobile device 18, utilize the microphone capability of mobile device 18 to listen and communicate with the background.

[0071] The backend 20 provides the analysis process for automatic audio measurement. The backend 20 includes a source automatic measurement signature generation module 28, a data visualization module 26, a storage module 36, an image similarity detection module 24, and an evaluation and transmission module 22. The mobile device 18 communicates directly with the backend 20, and also communicates with the vehicle 12 and the backend 20 through the information exchange infrastructure 14.

[0072] The automatic measurement signature generation module 28 includes a set of submodules 50, 52, and 54. Submodule 50 uses an algorithm mapping of audio frequency, character, and character position to encode the source data; submodule 52 determines the timing and duration of simultaneous tones; and submodule 54 encodes the audio file and stores it in a known automatic measurement signature container. Information from the automatic measurement signature generation module 28 is sent to the data visualization module 26 and the storage module 36.

[0073] The data visualization module 26 includes a set of sub-modules 40, 42, and 46. Sub-module 40 filters, analyzes, and represents automatically measured signal data that is biased for frequency and biased for time signature; sub-module 42 generates an automatically measured vehicle audio signature depiction as sample data 44; and using the sample data 44, sub-module 46 generates a color image 48 by removing unused frequencies and mapping the used frequencies to a predetermined RGB color map. Figure 2 The amplitude and duration of the detected 16 individual tones are shown. Further details of the analysis process associated with background 20 are described below.

[0074] System 10 encodes and stores exclusive attributes (metadata) of vehicles, players, and using combinations of audio and color to create vehicle / object-specific audio signature messages and corresponding encoded message image representations. System 10 uses selected frequency ranges to create equivalents of musical chords (simultaneously played tones) for each data segment, such as... Figure 3 As shown in the image. Each chord provides the vehicle's object ID, vehicle type, and vehicle trim level.

[0075] The sequential execution of chords is enhanced by purposeful pauses (mutes) between chord broadcasts. These pauses are applied in alternating chord-mute sequences. This sequence establishes rhythm (chord and mute durations, such as...) Figure 4 (As shown in the diagram). Rhythm is applied using mathematical functions. For example, given an initial minimum duration x (in milliseconds), a duration interval value y (in milliseconds), and a character position z (the character's position in the character set), the expression d (duration) = x + (y * z) – y. A unique (vehicle- or object-specific) sequence of chords and silences is also represented as a synesthetic image depicting the frequency and temporal aspects of conveying the encoded message. Therefore, the automatic measurement is a unique combination of frequency chords and associated rhythms used to transmit data.

[0076] Return to reference Figure 1The source automatic measurement signature generation module 28 provides character-to-frequency mapping and character-to-time mapping. For character-to-frequency mapping, the first and last chords in each transmission are predefined sets of synchronized tones outside the character mapping range. The content string is encoded into chords, and a different audio signature is given to each chord in the defined character set. Each character of the content string is represented by a frequency value (Hz) using the following expression:

[0077] Hz=(hz_minimum+((position_in_string*character_set_length*hz_delta)+((character_number+1)*hz_delta))-hz_delta).

[0078] For the character-to-time mapping, the duration of each chord and the inserted silence are derived by the following expression: ms = ms_minimum + (position * ms_delta) – ms_delta. And for the character-to-color mapping, hue, saturation, and luminance (HSL) color values ​​are assigned to each character based on the character and its position in the parent content string. Data 38 from the source automatic measurement signature generation module 28 is sent to the data visualization module 26. Data from the source automatic measurement signature generation module 28 is also sent to the storage module 36 as a known automatic measurement signature.

[0079] The data visualization module 26 encodes the data 38. Specifically, the data visualization module 26 provides audio signature analysis and signal visualization, starting with the audio file 38. The signal data visualization provides a spectrogram image and a color image depiction. The spectrogram image is data processed to generate a black-and-white spectrogram image representing the complete time and frequency domains, and the color image depiction is data processed to generate a compressed color image.

[0080] Data from the data visualization module 26 is sent to the determination module 34 to identify the source of the audio data. If the source is from the automatic measurement signature generation module 28, the data from the data visualization module is transferred to the storage module 36. Vehicle signature messages, vehicle attributes, vehicle black-and-white spectrograms, and vehicle color spectrograms are sent to a distributed data mining storage library. Image and audio files are distributed (shared) in the data mining storage library, which has the ability to increase its space footprint as the amount of data increases to reduce the data density per physical resource.

[0081] Next, the audio file is sent to the evaluation and transmission module 22. Here, the audio signature message is sent to the vehicle's internal audio module as an over-the-air (OTA) file push for broadcasting. Furthermore, the conditions for audio broadcasting are enabled via a trigger in the vehicle's internal module. For example, commercial use might require enabling interaction from the vehicle operator. Thus, the vehicle 12 plays the audio signature message using the infotainment or external speaker system.

[0082] Next, mobile device 18 provides detection and interaction with the cloud. Through mobile device 18, an indicator is sent to the user to confirm that the microphone is active and recording, and to initiate audio recording. Feedback to the user is provided via a two-factor interaction model for successful detection and failed confirmations (via SMS / MMS / Push).

[0083] In addition, the mobile device 18 sends information OTA to the backend 20, specifically the data visualization module 26. In various devices, this information may include encapsulated applicable metadata and audio files, including, for example, gamified sending accounts, gamified destination accounts, emergency sending accounts, and emergency destination accounts.

[0084] If the decision module 34 determines that the data from the data visualization module 26 originates from the mobile device 18, the data is transmitted to the image similarity detection module 24, where a set of images 30 undergoes deep learning image comparison to perform analysis matching 32.

[0085] Module 24 analyzes the RGB content of the receiver's compressed colorized image to identify sync and end chords, thereby aligning the message start and end points. Module 24 also performs a pixel-by-pixel comparison of the compressed color image stored in module 36. The output of module 24 is the percentage of pixel matching distribution, and a successful match is based on a tolerance standard.

[0086] As described above, Module 24 also provides advanced AI deep learning. Therefore, a decision model is generated using the dataset and known results, which is then applied to new data for judgment. Subsidiary applications also utilize this model to increase performance confidence. When new data becomes available, the model is retrained and improved, and then applied to the new data for judgment. Additionally, Module 24 performs group comparison size reduction using image analysis. Therefore, feature extraction is used to limit images to members of a subgroup for deep learning comparison. Subgroup approximation is performed by the deep learning system to identify confidence level tolerances, thereby addressing potential uncertainties derived from broadcast attributes. Information about the discovered image matches is sent (SMS / MMS / Push) back to the mobile device 18. Information can also be sent to other commercial systems, such as, for example, web service APIs.

[0087] System 10 provides one or more of the following benefits: enabling gaming activities involving the collection of vehicle broadcasts in a wide outdoor area to receive interactive feedback supported by a mobile application detection agency; providing communication of vehicle metadata; providing a usable detection device with an application that utilizes microphone capabilities to listen to and transmit captured audio and receive decoded signed messages and metadata; and providing a method for detecting broadcast metadata of law enforcement vehicles following suspicious vehicles or emergency first responders identifying distressed vehicles within physical distance.

[0088] The description in this disclosure is merely exemplary in nature, and changes that do not depart from the spirit and scope of this disclosure are intended to be within its scope. Such changes should not be considered as a departure from the spirit and scope of this disclosure.

Claims

1. A method for motor vehicle detection and identification through audio propagation, the method comprising: sending, by the motor vehicle, a uniquely identifiable audio signal; listening, by a mobile device, to the audio signal and sending the audio signal to a data visualization module; generating, by an automatic metric signature generation module, an automatic metric signature and sending the automatic metric signature to the data visualization module; broadcasting, by the motor vehicle, an automatic metric signature; detecting, by the mobile device, the automatic metric signature and sending the automatic metric signature to the data visualization module; visualizing, by the data visualization module, the audio signal and the automatic metric signature to provide image data and sending the image data to a decision module; determining, by the decision module, the source of the image data; if the source of the image data is determined to be a known automatic metric signature, evaluating and sending an automatic metric signature to the motor vehicle; and if the source of the image data is determined to originate from the mobile device, detecting image similarity of the signature with deep learning image comparison and sending information of the found image match back to the mobile device.

2. The method of claim 1, wherein the motor vehicle is in communication with an information exchange infrastructure.

3. The method of claim 2, wherein the mobile device is in communication with the information exchange infrastructure.

4. The method of claim 1, wherein generating an automatic metric signature comprises a character to frequency mapping.

5. The method of claim 1, wherein generating an automatic metric signature comprises a character to time mapping.

6. The method of claim 1, wherein generating an automatic metric signature comprises a character to color mapping.

7. The method of claim 1, wherein visualizing the automatic metric signature comprises audio signature analysis and signal data visualization.

8. The method of claim 1, further comprising storing data associated with the automatic metric signature in a source repository.

9. A method for motor vehicle detection and identification through audio propagation, the method comprising: sending, by the motor vehicle, a uniquely identifiable audio signal; listening, by a mobile device, to the audio signal and sending the audio signal to a data visualization module; generating, by an automatic metric signature generation module, an automatic metric signature comprising a character to frequency mapping, a character to time mapping, and a character to color mapping and sending the automatic metric signature to the data visualization module; broadcasting, by the motor vehicle, an automatic metric signature; detecting, by the mobile device, the automatic metric signature and sending the automatic metric signature to the data visualization module; visualizing, by the data visualization module, the audio signal and the automatic metric signature comprising audio signature analysis and signal data visualization to provide image data and sending the image data to a decision module; determining, by the decision module, the source of the image data; if the source of the image data is determined to be a known automatic metric signature, evaluating and sending an automatic metric signature to the motor vehicle; and if the source of the image data is determined to originate from the mobile device, detecting image similarity of the signature with deep learning image comparison and sending information of the found image match back to the mobile device. If it is determined that the source of the image data originates from the mobile device, then a deep learning image comparison is utilized to detect the signed image similarity and information of the found image match is sent back to the mobile device.

10. The method of claim 9, wherein the motor vehicle is in communication with an information exchange infrastructure.

11. The method of claim 10, wherein, The mobile device is in communication with the information exchange infrastructure.

12. The method of claim 9, further comprising storing data associated with the autometric signature in a source repository.

13. The method of claim 9, further comprising evaluating a sequence of chords to determine a distinct audio signature.

Images