Dynamic correlation signature generation

Abstract

To provide an efficient classification system and method in which one of main tasks related to driving is classification.SOLUTION: Wherein a first signature generated in a first iteration comprises a first identifier indicative of at least one of a feature of a road element associated with the first signature or a generated feature of the first signature; The first identifiers are generated in association with each other, and a second signature generated in a second iteration process includes a second identifier indicating a feature of a road element associated with the second signature or a generated feature of the second signature; The second identifiers are generated in association with each other, wherein a second identifier of the second signature is generated independently of a first identifier of the first signature, and wherein the first signature and the second signature collectively represent a sensing information cluster.SELECTED DRAWING: Figure 1

Description

[Background technology]

[0001] Driver assistance and automated driving systems are known in the art. In such systems, a computer-implemented system controls (at least to some extent) some or all of the vehicle's driving functions, such as speed, telemetry, braking, etc. The vehicle typically includes one or more sensors to provide the system with current information about the driving environment. The driving system typically uses the current information about the driving environment to determine how to drive on a road.

[0002] One of the main tasks involved in driving is classification.

[0003] There is an increasing need to provide efficient classification systems and methods. Summary of the Invention

[0004] SUMMARY OF THE INVENTION A method, system, and non-transitory computer readable medium are presented herein. [Brief explanation of the drawings]

[0005] Embodiments of the present disclosed subject matter will be more fully understood and appreciated from the following detailed description taken in conjunction with the drawings, in which: [Figure 1] 1 is an example of a system. [Figure 2] 1 is an example of a system. [Figure 3] 1 is an example of a method. [Figure 4] 1 is an example of a method. [Figure 5] 1 is an example of an identifier and signature set. [Figure 6] 1 is an example of a method. [Figure 7] 1 is an example of intra-signature relationships and intra-signature irrelevance. DETAILED DESCRIPTION OF THE INVENTION

[0006] A method, system, and computer-readable medium are provided.

[0007] The proposed solution provides accurate and robust signatures that enable the detection of road elements (road objects and / or road scenes).

[0008] The proposed solution takes into account the associations between signature elements (e.g., identifiers) and allows associating more than one signature per road element, significantly improving the accuracy of the classification.

[0009] According to an embodiment, there is provided a signature generated at least in part using a neural network, the signature including an identifier.

[0010] According to an embodiment, the identifier indicates at least one of (a) a characteristic of the road element associated with the first signature, or (b) a generated characteristic of the first signature.

[0011] According to an embodiment, the identifier represents the non-zero bits of a sparse representation of a neural network feature vector.

[0012] According to an embodiment, said identifiers represent activated neurons of a neural network.

[0013] According to an embodiment, the identifier represents non-zero bits of a sparse representation of a neural network feature vector. Different bits are associated with different properties. The properties may be selected in any manner and may be similar to embedding properties. The identifier may be a pointer to the non-zero bits.

[0014] According to an embodiment, the identifiers represent activated neurons (part of the activated neurons or all of the activated neurons) of the neural network, which may be neurons having the most impactful response to the sensory information units fed to the neural network.

[0015] According to an embodiment, there is provided a computer-implemented method for use in generating a dynamic correlation signature, the method comprising: A. generating a first signature in a first iteration, the first signature including a first identifier indicating at least one of (a) a road element characteristic associated with the first signature, or (b) the generated characteristics of the first signature, the first identifiers being generated in association with each other, and determining the first identifier for each iteration of the first iteration based on a relative occurrence of the first identifier with respect to a corresponding first true positive signature set and a corresponding first false positive signature set, so that for each iteration of the first iteration, the first true positive signature set and the first false positive signature set are determined based on the previous first true positive signature set and the previous first false positive signature set, respectively; B. generating a second signature in a second iterative process, the second signature including (a) a feature of a road element associated with the second signature, or (b) a second identifier indicating the generated feature of the second signature, the second identifiers being generated in association with each other, and determining the second identifier for each iteration of the second iterative process based on the relative occurrence of the second identifier with respect to a corresponding second true positive signature set and a corresponding second false positive signature set, so that for each iteration of the second iterative process, the second true positive signature set and the second false positive signature set are determined based on the previous second true positive signature set and the previous second false positive signature set. The second identifier of the second signature is generated independently of the first identifier of the first signature, and the first signature and the second signature collectively represent a sensed information cluster.

[0016] According to an embodiment, the method includes generating a third signature in a third iteration, the third signature including (a) a road element characteristic associated with the third signature, or (b) a third identifier indicating the generated characteristic of the third signature, the third identifiers being generated in association with each other, and determining, for each iteration of the third iteration, a relative occurrence of the third identifier in a corresponding third true positive signature set and a corresponding third false positive signature set. By determining the third identifier based on the current situation, for each iteration of the third iterative process, a third true positive signature set and a third false positive signature set are determined based on the previous third true positive signature set and the previous third false positive signature set, wherein the third identifier of the third signature is generated independently of the second identifier of the second signature, and the second signature and the third signature collectively represent a sensed information cluster.

[0017] According to an embodiment, the first signature represents the content of a plurality of sensory information units.

[0018] According to an embodiment, the first identifier represents non-zero bits of a sparse representation of a neural network feature vector.

[0019] According to an embodiment, the first identifier represents an activated neuron of a neural network.

[0020] According to an embodiment, the method includes generating additional signatures until convergence is reached.

[0021] According to an embodiment, the first signature and the second signature relate to road elements represented by the acquired information.

[0022] According to an embodiment, the first set of false positive signatures represents road elements similar to the road element.

[0023] According to an embodiment, the method includes generating additional signatures to provide a signature cluster associated with the road element and granting an inference process access to the cluster.

[0024] According to an embodiment, for one of a plurality of iterations in the first iterative process, the first set of true positive signatures and the first set of false positive signatures are determined in relation to perception data.

[0025] According to an embodiment, for one of a plurality of iterations in the first iterative process, the first set of true positive signatures and the first set of false positive signatures are determined in relation to sensed information.

[0026] According to an embodiment, for one of the iterations of the first iteration process, the first set of true positive signatures and the first set of false positive signatures are determined in relation to a target object.

[0027] According to an embodiment, the method includes analyzing the first generated signature and the second generated signature in a real-time application.

[0028] According to an embodiment, the method includes providing the first generated signature and the second generated signature to an offline application and analyzing the first signature and the second signature in the offline application.

[0029] According to an embodiment, a non-transitory computer-readable medium for use in generating a dynamic correlation signature is provided, the non-transitory computer-readable medium storing instructions that, when executed by a processing circuit, cause the processing circuit to generate a first signature in a first iteration, the first signature including a first identifier indicating at least one of (a) a road element characteristic associated with the first signature or (b) a generated characteristic of the first signature, the first identifiers being generated in association with each other, and determining the first identifier for each iteration of the first iteration based on a relative occurrence of the first identifier with respect to a corresponding first true positive signature set and a corresponding first false positive signature set, so that for each iteration of the first iteration, the first true positive signature set and the first false positive signature set are determined based on a previous first true positive signature set and a previous first false positive signature set, respectively. and (ii) generating a second signature in a second iteration, the second signature including a second identifier indicating (a) a road element characteristic associated with the second signature or (b) a generated characteristic of the second signature, the second identifiers being generated in association with each other, and determining, for each iteration of the second iteration, a relative occurrence status of the second identifier with respect to a corresponding second true positive signature set and a corresponding second false positive signature set. The second identifier is determined based on the situation, and for each iteration of the second iterative process, a second set of true positive signatures and a second set of false positive signatures are determined based on the previous second set of true positive signatures and the previous second set of false positive signatures, wherein the second identifier of the second signature is generated independently of the first identifier of the first signature, and the first signature and the second signature collectively represent a sensed information cluster.

[0030] According to an embodiment, the computer-readable medium stores instructions, the instructions being used to execute generating a third signature in a third iteration, the third signature including (a) a road element characteristic associated with the third signature, or (b) a third identifier indicating the generated characteristic of the third signature, the third identifiers being generated in association with each other, and a corresponding third true positive signature set and a corresponding third false positive signature set for each iteration of the third iteration. By determining the third identifier based on the relative occurrence of the third identifier relative to the previous third true positive signature set and the previous third false positive signature set, for each iteration of the third iterative process, a third true positive signature set and a third false positive signature set are determined based on the previous third true positive signature set and the previous third false positive signature set, where the third identifier of the third signature is generated independently of the second identifier of the second signature, and the second signature and the third signature collectively represent a sensed information cluster.

[0031] According to an embodiment, the first signature represents the content of a plurality of sensory information units.

[0032] According to an embodiment, the first identifier represents non-zero bits of a sparse representation of a neural network feature vector.

[0033] According to an embodiment, the first identifier represents an activated neuron of a neural network.

[0034] According to an embodiment, the computer readable medium stores instructions for generating additional signatures until convergence is reached.

[0035] The different figures show examples of units and / or software and / or information items and / or steps and / or components. Said examples are provided for ease of interpretation. At least one of the units and / or software and / or information items and / or steps and / or components is optional or mandatory.

[0036] 1 shows an example of a computerized system 100, which includes a communication system 130, one or more memory and / or storage units 120, and a processing system 124 including a processor 126. The computerized system may be a server, a laptop computer, a desktop computer, or any other computer, and may include or be in communication with sensing units and / or controllers.

[0037] According to an embodiment, computerized system 100 is in communication with a network 132 and one or more other remote computerized systems 134 , which in turn are in communication with network 132 .

[0038] According to an embodiment, the communication system 130 is configured to enable communication between one or more memory and / or storage units 120 and / or the sensing system 110 and / or any one of the additional units and / or the network 132 (which communicates with remote computerized systems).

[0039] The memory and / or storage unit 120 is shown as storing software, and any reference to software should apply mutatis mutandis to code and / or firmware and / or instructions and / or commands, etc.

[0040] Processor 126 includes multiple processing units 126(1) through 126(J), where J is an integer greater than 1. Any reference to a unit or item should apply mutatis mutandis to multiple units or items. For example, any reference to a processor should apply mutatis mutandis to multiple processors, and any reference to communication system 130 should apply mutatis mutandis to multiple communication systems.

[0041] According to an embodiment, the one or more memory and / or storage units 120 may include one or more memory units, each of which may include one or more memory banks.

[0042] According to an embodiment, the one or more memory and / or storage units 120 include volatile memory and / or non-volatile memory. The one or more memory and / or storage units 120 may be random access memory (RAM) and / or read-only memory (ROM).

[0043] According to an embodiment, the non-volatile memory unit is a mass storage device that can provide non-volatile storage of computer code, computer-readable instructions, data structures, program modules, and other data for a processor or any other unit of the vehicle. For example, but not limited to, the mass storage device can be a hard disk, a removable magnetic disk, a removable optical disk, a magnetic tape cassette or other magnetic storage device, a flash memory card, a CD-ROM, a digital versatile disk (DVD) or other optical storage device, a random access memory (RAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), etc.

[0044] Any content may be stored in any part or type of memory and / or storage unit.

[0045] According to an embodiment, the at least one memory unit stores at least one database, such as any database known in the art, such as DB2®, Microsoft® Access, Microsoft® SQL Server, Oracle®, mySQL, PostgreSQL, etc.

[0046] The various units and / or components communicate with each other using any communication element and / or protocol. An example of a communication system is represented at 130. Other communication elements may also be provided.

[0047] Communications system 130 can communicate with bus 136. The bus represents one or more of several possible types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, and a processor or local bus using any one of a variety of bus architectures. By way of example, such architectures may include an Industry Standard Architecture (ISA) bus, a MicroChannel Architecture (MCA) bus, an Enhanced ISA (EISA) bus, a Video Electronics Standards Association (VESA) local bus, an Accelerated Graphics Port (AGP) bus and a Peripheral Component Interconnect (PCI), a PCI-Express bus, a Personal Computer Memory Card International Association (PCMCIA), a Universal Serial Bus (USB), etc. The bus, and all buses specified herein, may be implemented by wired or wireless network connections and respective subsystems of the subsystems.

[0048] Network 132 is located outside the vehicle and is used for communication between the vehicle and at least one remote computing system. By way of example, the remote computing system may be a personal computer, laptop computer, portable computer, server, router, network computer, peer-to-peer device, or other common network node. Logical connections between the processor and any one of the remote computing systems may be made via local area networks (LANs) and general wide area networks (WANs). Such network connections may be made by network adapters (which may belong to communication system 130), which may be implemented in both wired and wireless environments. Such network environments are common in offices, enterprise-wide computer networks, and larger networks such as intranets and the Internet.

[0049] It should be noted that at least a portion of the content shown to be stored in one or more memory / storage units 120 may be stored external to the vehicle. It should also be noted that the processor may evaluate signatures generated by multiple detectors.

[0050] According to an embodiment, the memory and / or storage unit 120 stores at least one of an operating system 194, acquired information 181, a true positive signature set 182, a false positive signature set 183, signatures 184 generated in an iterative process (using the true positive signature set 182 and the false positive signature set 183), true positive / false positive (TP / FP) generation software 185 for generating at least a portion of the true positive signature set 182 and the false positive signature set 183, signature generation software 186, identifier generation software 187, and additional software 188.

[0051] Using software, the processing system is configured to perform method 200 and / or step 300 and / or method 800.

[0052] The computerized system may be located anywhere, for example, inside the vehicle, between the vehicles, outside the vehicle, etc.

[0053] FIG. 2 shows an example of a vehicle 400 , a network 432 and a remote computerized system 434 .

[0054] The vehicle 400 includes a sensing system 410, a communication system 430, one or more memory and / or storage units 420, and additional units including a control unit 425, an advanced driver assistance system (ADAS) control unit 423, an autonomous driving control unit 422, and a processing system 424 including a processor 426. The network 423 communicates with the vehicle and with remote computerized systems 434, such as servers, cloud computers, etc.

[0055] The communication system 430, the one or more memory and / or storage units 420, and the processing system 434 may form a computerized system that may include one or more other systems and / or units, such as the sensing system 410.

[0056] The communication system 430 is configured to facilitate communication between one or more memory and / or storage units 420, and / or the sensing system 410, and / or any one of the additional units, and / or a network 432 (for communicating with remote computerized systems).

[0057] The control unit 425 may interface with an advanced driver assistance system (ADAS) control unit 423, an autonomous driving control unit 422, and / or may control or communicate with other vehicle components (including the vehicle computer 421).

[0058] The one or more memory and / or storage units 420 are shown to store an operating system 494, software 493 (particularly software necessary for performing perception tasks and / or classification tasks and / or tasks affecting vehicle driving operations), clusters 481 generated by executing method 200 and / or method 800, classifier software 482 for classifying road elements sensed during vehicle driving, perception software 483 for providing and processing information about the environment around the vehicle, and sensory information 484 sensed by sensing system 410.

[0059] The sensing system 410 may include an optical device, a sensing element group, a readout circuit, and an image signal processor. After the optical device is a sensing element group, for example a sensing element row or a sensing element array forming the sensing element group. After the sensing element group is a readout circuit that reads the detection signals generated by the sensing element group. The image signal processor is configured to perform initial processing on the detection signals, for example improving the quality of the detection information, reducing noise, etc. The sensing system 410 is configured to output one or more sensing information units (SIUs).

[0060] The communication system 430 is configured to facilitate communication between one or more memory and / or storage units 420, and / or the sensing system 410, and / or any one of the additional units, and / or a network 432 (which communicates with a remote computerized system 434, which may include the computerized system 400).

[0061] The controller 425 is configured to control the operation of the sensing system 410, and / or one or more memory and / or storage units 420, and / or one or more additional units (other than the controller).

[0062] The ADAS control unit 423 is configured to control the ADAS operation.

[0063] The autonomous driving control unit 422 is configured to control the autonomous driving of the autonomous vehicle.

[0064] The vehicle computer 421 is configured to control the operation of the vehicle, and in particular to control the engine, transmission and any other vehicle systems or components.

[0065] Processing system 424 may include processor 426 and one or more other processors and is configured to perform any of the methods described herein.

[0066] The one or more memory and / or storage units 420 are configured to store firmware and / or software, one or more operating systems, data and metadata necessary to perform any of the methods mentioned herein.

[0067] The vehicle computer 421 may communicate with an engine control module, a transmission control module, a powertrain control module, and the like.

[0068] The memory and / or storage unit 420 is shown as storing software, and any reference to software should apply mutatis mutandis to code and / or firmware and / or instructions and / or commands, etc.

[0069] Processor 426 includes multiple processing units 426(1) through 426(J), where J is an integer greater than 1. Any reference to a unit or item should apply mutatis mutandis to multiple units or items. For example, any reference to a processor should apply to multiple processors, and any reference to communication system 430 should apply mutatis mutandis to multiple communication systems.

[0070] According to an embodiment, the one or more memory and / or storage units 420 may include one or more memory units, each of which may include one or more memory banks.

[0071] According to an embodiment, the one or more memory and / or storage units 420 include volatile memory and / or non-volatile memory. The one or more memory and / or storage units 420 may be random access memory (RAM) and / or read-only memory (ROM).

[0072] According to an embodiment, the non-volatile memory unit is a mass storage device that can provide non-volatile storage of computer code, computer-readable instructions, data structures, program modules, and other data for a processor or any other unit of a vehicle. For example, but not limited to, the mass storage device can be a hard disk, a removable magnetic disk, a removable optical disk, a magnetic tape cassette or other magnetic storage device, a flash memory card, a CD-ROM, a digital versatile disk (DVD) or other optical storage device, a random access memory (RAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), etc.

[0073] Any content may be stored in any part or type of memory and / or storage unit.

[0074] According to an embodiment, the at least one memory unit stores at least one database, such as any database known in the art, such as DB2®, Microsoft® Access, Microsoft® SQL Server, Oracle®, mySQL, PostgreSQL, etc.

[0075] The various units and / or components communicate with each other using any communication element and / or protocol. An example of a communication system is represented at 430. Other communication elements may also be provided.

[0076] FIG. 2 shows a communication system 430 in communication with various processors and / or units and a network 432 .

[0077] The communications system 430 may include a bus. The bus represents one or more of several possible types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, and a processor or local bus using any one of a variety of bus architectures. By way of example, such architectures may include an Industry Standard Architecture (ISA) bus, a MicroChannel Architecture (MCA) bus, an Enhanced ISA (EISA) bus, a Video Electronics Standards Association (VESA) local bus, an Accelerated Graphics Port (AGP) bus, a Peripheral Component Interconnect (PCI), a PCI-Express bus, a Personal Computer Memory Card International Association (PCMCIA), a Universal Serial Bus (USB), etc. The bus, and all buses specified herein, may be implemented by wired or wireless network connections and subsystems, respectively.

[0078] Network 432 is located outside the vehicle and is used for communication between the vehicle and at least one remote computing system. By way of example, the remote computing system may be a personal computer, laptop computer, portable computer, server, router, network computer, peer-to-peer device, or other common network node. Logical connections between the processor and any one of the remote computing systems may be made via local area networks (LANs) and general wide area networks (WANs). Such network connections may be made by network adapters (which may belong to communication system 430), which may be implemented in both wired and wireless environments. Such network environments are common in offices, enterprise-wide computer networks, and larger networks such as intranets and the Internet.

[0079] It should be noted that at least a portion of the content shown to be stored in one or more memory / storage units 420 may be stored external to the vehicle. It should also be noted that the processor may evaluate signatures generated by multiple detectors.

[0080] According to an embodiment, the processor is configured as described above.

[0081] FIG. 3 illustrates an example computer-implemented method 200 used to generate dynamic correlation signatures.

[0082] According to an embodiment, the method 200 includes an initialization step 205 .

[0083] According to an embodiment, the initialization step 205 includes at least one of the following items: A. Obtain an initial first set of true-positive signatures and an initial first set of false-positive signatures to be used in the first iteration of the first iteration process. B. Obtain an initial signature of a road element, and generate an initial first set of true positive signatures and an initial first set of false positive signatures based on the initial signature.

[0084] According to an embodiment, the initialization step is followed by step 210, which generates a first signature associated with the acquired information. The first signature includes a first identifier, which is determined by creating a first set of true positive signatures and a first set of false positive signatures in a first iteration. Each first set of true positive signatures and each first set of false positive signatures in a corresponding iteration are determined in relation to a previous first set of true positive signatures and a previous set of first false positive signatures in an iteration prior to the first iteration.

[0085] According to an embodiment, each first identifier in a corresponding iteration is determined based on the relative occurrence of the first identifier with respect to the first set of true positive signatures and the first set of false positive signatures in the corresponding iteration.

[0086] According to an embodiment, each first identifier in a corresponding iteration is determined based on a trade-off between the relative occurrence of the first identifier in the corresponding iteration with respect to the first true positive signature set and the relative occurrence of the first identifier in the corresponding iteration with respect to the first false positive signature set.

[0087] According to an embodiment, the first true positive signature set in a corresponding iteration is included in the previous first true positive signature set in a previous iteration.

[0088] According to an embodiment, another first set of true positive signatures in another corresponding iteration only partially overlaps with another previous first set of true positive signatures in another previous iteration.

[0089] According to an embodiment, step 210 is completed when convergence of the first iteration is reached. Examples of such convergence include finding that (i) the first set of true positive signatures and the first set of false positive signatures in a corresponding iteration are approximately equal to (ii) the first set of true positive signatures and the first set of false positive signatures in an iteration subsequent to the corresponding iteration.

[0090] According to an embodiment, step 210 is followed by step 220, which generates a second signature including a second identifier determined by creating a second set of true positive signatures and a second set of false positive signatures in a second iteration, where each second set of true positive signatures and each second set of false positive signatures in a corresponding iteration are determined in association with a previous set of true positive signatures and a previous set of false positive signatures in a previous iteration of the second iteration, thereby generating the second signature independently of the first signature.

[0091] According to an embodiment, the method 200 includes obtaining an initial first true positive signature set and an initial first false positive signature set associated with a first iteration of a first iteration process, and the first identifier is associated with the first signature true positive signature subset and the first signature first false positive signature subset.

[0092] According to an embodiment, step 220 includes (i) generating an initial second true positive signature set associated with the first iteration of the second iteration by removing from the initial first true positive signature set any signatures that match the first signature (matching may include at least a predetermined number of identifiers appearing in the first signature and the initial first true positive signature), and (ii) generating an initial second false positive signature set associated with the first iteration of the second iteration by removing from the initial first false positive signature set any signatures that match the first signature (matching may include at least a predetermined number, or at least a predetermined percentage, for example, at least 60, 65, 70, 75, 80, 85, 90, 95 percent, of identifiers appearing in the first signature and the initial first false positive signature).

[0093] Said removal, when performed each time a new signature is generated, renders different signatures irrelevant.

[0094] Although steps 210 and 220 involve generating a first signature and a second signature, according to an embodiment, method 200 is performed to generate one or more additional signatures, each of which is generated independently of the previous signature.

[0095] According to an embodiment, the method includes generating additional signatures until convergence is reached. An example of convergence may include associating all or a majority of the signatures defined in the initial first true-positive signature set and the initial first false-positive signature set with the signatures generated in method 200. The generation of the additional signatures is described below in step 230. Step 230 generates additional signatures including additional identifiers determined by creating additional true-positive signature sets and additional false-positive signature sets in additional iterations. Each additional true-positive signature set and each additional false-positive signature set in a corresponding iteration is determined in association with a previous additional true-positive signature set and a previous additional false-positive signature set in a previous iteration of the additional iteration, thereby generating the additional signatures independently of any of the previously generated signatures.

[0096] According to an embodiment, step 230 is followed by step 240. Step 240 determines whether to generate another additional signature, if yes, proceed to step 230, if no, end signature generation.

[0097] According to an embodiment, the signatures generated by the method 200 form signature clusters associated with road elements.

[0098] The method 200 may be repeated in association with different road elements to provide multiple signature clusters associated with different road elements.

[0099] According to an embodiment, the method 200 includes a step 250 of responding to the completion of the generation of the signature.

[0100] Step 250 may include at least one of the following items: A. The clusters formed by the generated signatures are entered into a database. B. Transmitting the generated signature to one or more vehicles associated with the road element to which the generated signature relates. C. Granting access to one or more clusters generated during execution of method 200. The access may be provided to an inference process for classifying road elements sensed by the vehicle. D. determining whether to repeat the method or another road element, and selectively repeating the method based on said determination. E. Analyzing the first generated signature and the second generated signature in a real-time application. Analysis may include checking whether the signature is accurate, e.g., whether the signature accurately represents the captured item, e.g., whether the signature is a true positive, true negative, false positive, or false negative. The real-time application may be related to vehicle operation. F. Providing the first generated signature and the second generated signature to an offline application, where the offline application analyzes the first signature and the second signature. Analysis may include checking whether the signature is accurate, e.g., whether the signature accurately represents the captured item, e.g., whether the signature is a true positive, true negative, false positive, or false negative. The real-time application may be related to vehicle operation.

[0101] According to an embodiment, the first set of false positive signatures represents other road elements similar to the road element, where similarity means that the other road elements may be mistaken for the road element of interest. Similarity can be determined manually or automatically, for example, by analyzing false positive signatures of road elements that are different from the road element of interest but have been mistakenly identified as the road element of interest.

[0102] According to an embodiment, any signature is based on the cropped sensory information unit. U.S. patent application Ser. No. 18 / 595,368, filed Mar. 3, 2024, discloses a method for generating a signature based on a cropped image, and is incorporated herein by reference.

[0103] 4 includes a step 300 of generating a signature. Step 300 is an example of any one of steps 210, 200, or 230 in method 200.

[0104] According to an embodiment, step 300 includes the following series of steps: 302, 304, 306, and 308.

[0105] The index n indicates the order of the signature and the order of the iterations applied during the generation of the signature.

[0106] The index k indicates the order in which the identifiers are identified.

[0107] According to an embodiment, step 302 includes obtaining an initial n-th set of true positive signatures and an initial n-th set of false positive signatures. If n is equal to 1, step 302 is linked to step 205 of method 200. If n is greater than 1, step 302 includes obtaining an initial set of signatures that is independent of the generation of one or more previous signatures.

[0108] According to an embodiment, step 304 includes determining a first identifier in the nth signature based on (i) the relative occurrence of said first identifier in the initial nth set of true positive signatures and (ii) the relative occurrence of said first identifier in the initial nth set of false positive signatures. Step 304 provides a trade-off between (i) and (ii), which may be a weighting and / or any other function aimed at selecting identifiers with a high relative occurrence in the initial nth set of true positive signatures and a low relative occurrence in the initial nth set of false positive signatures.

[0109] According to an embodiment, step 306 includes generating a (k,n)th true positive signature set and a corresponding (k,n)th false positive signature set relative to the previous nth true positive signature set and the previous nth false positive signature set in the previous iteration.

[0110] According to an embodiment, step 308 includes determining the (k,n)th first identifier in the nth signature based on the relative occurrence of the (k,n)th identifier in the (k,n)th true positive signature set and the (k,n)th false positive signature set.

[0111] Steps 306 and 308 (with index k having different values) may be repeated until signature generation is complete.

[0112] FIG. 5 illustrates an example of a set of true positive signatures associated with the generation of a first signature, a set of false positive signatures associated with the generation of a first signature, and an identifier associated with the generation of a first signature.

[0113] The true positive signature sets include initial, first, second, third, fourth and fifth true positive signature sets 910(0,1), 910(1,1), 910(2,1), 910(3,1), 910(4,1) and 910(5,1).

[0114] The false positive signature sets include initial, first, second, third, fourth and fifth false positive signature sets 920(0,1), 920(1,1), 920(2,1), 920(3,1), 920(4,1) and 920(5,1).

[0115] The first identifiers include 930(1,1), 930(2,1), 930(3,1), 930(4,1), and 930(5,1), which are the first, second, third, fourth, and fifth identifiers of the first signature.

[0116] According to the embodiment, A. The first TP signature set includes all signatures in the initial TP signature set that include the first signature's first identifier 930(1,1). B. The first FP signature set includes all signatures in the initial FP signature set that do not include the first signature's first identifier 930(1,1). C. The second TP signature set includes all signatures of the first TP signature set that include the second identifier 930(2,1) of the first signature. D. The second FP signature set includes all signatures in the first FP signature set that do not include the second identifier 930(2,1) of the first signature. E. The third TP signature set includes all signatures of the second TP signature set that include the third identifier 930(3,1) of the first signature. F. The third FP signature set includes all signatures in the second FP signature set that do not include the third identifier 930(3,1) of the first signature. G. The fourth TP signature set includes all signatures in the initial TP signature set that include at least half of the identifiers generated before the first signature (in this case, at least two of the first, second, and third identifiers of the first signature). H. The fourth FP signature set includes all signatures in the initial FP signature set that do not include at least half of the identifiers generated before the first signature (in this case, at least two of the first, second, and third identifiers of the first signature). I. The fifth TP signature set includes all signatures in the initial TP signature set that include at least half of the previously generated identifiers of the first signature (in this case, at least two of the first, second, third, and fourth identifiers of the first signature). J. The fifth FP signature set includes all signatures in the initial FP signature set that do not include at least half of the identifiers generated before the first signature (in this case, at least two of the first, second, third, and fourth identifiers of the first signature).

[0117] FIG. 6 illustrates an example computer-implemented method 800 used to generate dynamic correlation signatures.

[0118] According to an embodiment, the method 800 includes an initialization step 805 .

[0119] According to an embodiment, the initial step 805 includes at least one of the following items: A. Obtain an initial first set of true-positive signatures and an initial first set of false-positive signatures to be used in the first iteration of the first iteration process. B. Obtaining an initial signature of a road element, and generating the initial first set of true positive signatures and the initial first set of false positive signatures based on the initial signature.

[0120] According to an embodiment, step 805 is followed by step 810, which includes (i) generating a first signature in a first iteration, the first signature including a first identifier indicating at least one of (a) a road element feature associated with the first signature or (b) the generated features of the first signature, the first identifiers being generated in association with each other, and determining the first identifier for each iteration of the first iteration based on a relative occurrence of the first identifier with respect to a corresponding first true positive signature set and a corresponding first false positive signature set, whereby a first true positive signature set and a first false positive signature set are determined for each iteration of the first iteration based on the previous first true positive signature set and the previous first false positive signature set, respectively.

[0121] According to an embodiment, step 810 may be preceded by step 802 .

[0122] According to an embodiment, step 810 is followed by step 820. Step 820 generates a second signature in a second iteration, the second signature including (a) a road element feature associated with the second signature or (b) a second identifier indicating the generated feature of the second signature, the second identifiers being generated in association with each other, and determining the second identifier for each iteration of the second iteration based on a relative occurrence of the second identifier with respect to a corresponding second true positive signature set and a corresponding second false positive signature set, so that a second true positive signature set and a second false positive signature set are determined for each iteration of the second iteration based on the previous second true positive signature set and the previous second false positive signature set.

[0123] According to an embodiment, the second identifier of the second signature is generated independently of the first identifier of the first signature.

[0124] According to an embodiment, the first signature and the second signature collectively represent a cluster of sensed information.

[0125] According to an embodiment, step 830 generates an additional signature in an additional iterative process, the additional signature including (a) a feature of a road element associated with the additional signature, or (b) an additional identifier indicating the generated feature of the additional signature, the additional identifiers being generated in association with each other, and determining the additional identifier for each iteration of the additional iterative process based on the relative occurrence of the additional identifier with respect to a corresponding additional true positive signature set and a corresponding additional false positive signature set, so that for each iteration of the additional iterative process, an additional true positive signature set and an additional false positive signature set are determined based on the previous additional true positive signature set and the previous additional false positive signature set, wherein the additional identifier of the additional signature is generated independently of the second identifier of the second signature, and the second signature and the additional signature collectively represent a sensory information cluster. The additional signature may be the third signature or a signature after generating the third signature.

[0126] According to an embodiment, the method 800 may continue to generate additional information until convergence is reached.

[0127] According to an embodiment, step 830 is followed by step 840. Step 840 determines whether to generate another additional signature, if yes, proceed to step 830, if no, end signature generation.

[0128] According to an embodiment, once all signatures have been generated, step 840 transitions to step 850, which responds to the generation of the signatures.

[0129] According to an embodiment, step 850 may include at least one of the following items: A. The clusters formed by the generated signatures are entered into a database. B. Transmitting the generated signature to one or more vehicles associated with the road element to which the generated signature relates. C. Granting access rights to one or more clusters generated during execution of method 200. The access rights may be provided to an inference process for classifying road elements sensed by the vehicle. D. determining whether to repeat the method or another road element, and selectively repeating the method based on said determination. E. Analyzing the first generated signature and the second generated signature in a real-time application. Analysis may include checking whether the signature is accurate, e.g., whether the signature accurately represents the captured item, e.g., whether the signature is a true positive, true negative, false positive, or false negative. The real-time application may be related to vehicle operation. F. Providing the first generated signature and the second generated signature to an offline application, where the offline application analyzes the first signature and the second signature. Analysis may include checking whether the signature is accurate, e.g., whether the signature accurately represents the captured item, e.g., whether the signature is a true positive, true negative, false positive, or false negative. The real-time application may be related to vehicle operation.

[0130] According to an embodiment, the first signature represents the content of a plurality of sensory information units.

[0131] According to an embodiment, the first identifier represents non-zero bits of a sparse representation of a neural network feature vector, where different bits are associated with different properties. The properties may be selected in any manner and may be similar to embedding properties. The first identifier may be a pointer to the non-zero bits.

[0132] According to an embodiment, the first identifier represents activated neurons (some activated neurons or all activated neurons) of the neural network, which may be neurons having the most impactful response to the sensory information units fed to the neural network.

[0133] According to an embodiment, the first signature and the second signature relate to road elements represented by the acquired information.

[0134] According to an embodiment, the first set of false positive signatures represents road elements similar to the road element.

[0135] According to an embodiment, the method includes generating additional signatures to provide a signature cluster associated with the road element and granting an inference process access to the cluster.

[0136] According to an embodiment, for one of the iterations of the first iteration process, the first set of true positive signatures and the first set of false positive signatures are determined in relation to perception data.

[0137] According to an embodiment, for one of the iterations of the first iterative process, the first set of true positive signatures and the first set of false positive signatures are determined in relation to the sensory information.

[0138] According to an embodiment, for one of the iterations of the first iteration process, the first set of true positive signatures and the first set of false positive signatures are determined in relation to a target object.

[0139] By way of example, and not intended to be limiting, computer-readable media may include "computer storage media" and "communications media." "Computer storage media" includes volatile and non-volatile, removable and non-removable media, implemented in any method or technology, for storage of information such as computer-readable instructions, data structures, program modules, or other data. Exemplary computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVDs) or other optical storage devices, magnetic tape cassettes, magnetic tape, magnetic disk storage devices or other magnetic storage devices, or any other medium that can be used to store the desired information and that can be accessed by a computer.

[0140] According to an embodiment, the object classification system computer 701 is configured to perform any of the methods described herein.

[0141] According to an embodiment, the object classification system computer 701 communicates with one or more sensors of one or more types associated with the vehicle.

[0142] According to an embodiment, the object classification system computer 701 communicates with other vehicle computers, such as, for example, control computers (e.g., engine control computer, powertrain control computer) configured to control one or more vehicle units, and / or an autonomous driving unit configured to control autonomous driving, an ADAS unit configured to control ADAS operation, a route unit configured to navigate the vehicle, etc. Each unit includes processing circuitry and / or software and / or firmware and / or code and / or instructions stored on a non-transitory computer readable medium for completing the unit's role.

[0143] 7 shows examples of the following: (a) the relationship between the first to fifth identifiers (930(1,1) to 930(5,1)) of the first signature, (b) the relationship between the first to fifth identifiers (930(1,2) to 930(5,2)) of the second signature, and (c) the unrelationship between the identifiers of the first signature and the identifiers of the second signature.

[0144] Any combination of any steps of any method set forth herein is provided.

[0145] According to an embodiment, a computer-implemented method for generating dynamic correlation signatures is provided, the method including: (i) generating a first signature associated with acquired information, the first signature including a first identifier, the first identifier being determined by creating a first set of true positive signatures and a first set of false positive signatures in a first iteration, wherein each first set of true positive signatures and each first set of false positive signatures in a corresponding iteration are associated with a previous first set of true positive signatures and a previous set of first false positive signatures in a previous iteration of the first iteration. and (ii) generating a second signature, the second identifier including the second identifier, the second identifier being determined by creating a second set of true positive signatures and a second set of false positive signatures in a second iteration, wherein each second set of true positive signatures and each second set of false positive signatures in a corresponding iteration are determined in association with a previous set of true positive signatures and a previous set of false positive signatures in a previous iteration of the second iteration, thereby generating the second signature independently of the first signature.

[0146] According to an embodiment, each first identifier in a corresponding iteration is determined based on the relative occurrence of the first identifier with respect to the first set of true positive signatures and the first set of false positive signatures in the corresponding iteration.

[0147] According to an embodiment, each first identifier in a corresponding iteration is determined based on a trade-off between the relative occurrence of the first identifier in the first true positive signature set in the corresponding iteration and the relative occurrence of the first identifier in the first false positive signature set in the corresponding iteration.

[0148] According to an embodiment, the first true positive signature set in a corresponding iteration is included in the previous first true positive signature set in a previous iteration.

[0149] According to an embodiment, another first set of true positive signatures in another corresponding iteration only partially overlaps with another previous first set of true positive signatures in another previous iteration.

[0150] According to an embodiment, the method includes obtaining an initial first set of true-positive signatures and an initial first set of false-positive signatures for use in a first iteration period of a first iteration process.

[0151] According to an embodiment, the method includes obtaining an initial signature of a road element, and generating the initial first set of true positive signatures and the initial first set of false positive signatures based on the initial signature.

[0152] According to the claimed method, the method includes completing a first iteration process when (i) the first true positive signature set and the first false positive signature set in the corresponding iteration are approximately equal to (ii) the first true positive signature set and the first false positive signature set in the iteration after the corresponding iteration.

[0153] According to an embodiment, the method includes obtaining an initial first true positive signature set and an initial first false positive signature set associated with a first iteration of a first iteration process, wherein a first identifier is associated with the first signature true positive signature subset and the first signature first false positive signature subset.

[0154] According to an embodiment, the method includes: (i) generating an initial second true positive signature set associated with a first iteration of the second iteration by removing from the initial first true positive signature set any signatures that appear in the first signature true positive signature subset; and (ii) generating an initial second false positive signature set associated with the first iteration of the second iteration by removing from the initial first false positive signature set any signatures that appear in the first signature false positive signature subset.

[0155] According to an embodiment, the method includes generating additional signatures until convergence is reached.

[0156] According to an embodiment, the first signature and the second signature relate to road elements represented by the acquired information.

[0157] According to an embodiment, the first set of false positive signatures represents road elements similar to the road element.

[0158] According to an embodiment, the method includes generating additional signatures to provide a signature cluster associated with the road element and granting an inference process access to the cluster.

[0159] According to an embodiment, a non-transitory computer-readable medium for use in generating a dynamic correlation signature is provided, the non-transitory computer-readable medium storing instructions that, when executed by a processing circuit, cause the processing circuit to: (i) generate a first signature associated with acquired information, the first signature including a first identifier, the first identifier being determined by creating a first set of true positive signatures and a first set of false positive signatures in a first iteration, wherein each first set of true positive signatures and each first set of false positive signatures in a corresponding iteration are determined in relation to a previous first set of true positive signatures and a previous first set of false positive signatures in a previous iteration of the first iteration; (ii) generating a second signature including a second identifier, the second identifier being determined by creating a second set of true positive signatures and a second set of false positive signatures in a second iteration, wherein each second set of true positive signatures and each second set of false positive signatures in a corresponding iteration are determined in relation to a previous set of true positive signatures and a previous set of false positive signatures in a previous iteration of the second iteration; The second signature is generated independently of the first signature.

[0160] According to an embodiment, each first identifier in a corresponding iteration is determined based on the relative occurrence of the first identifier with respect to the first set of true positive signatures and the first set of false positive signatures in the corresponding iteration.

[0161] According to an embodiment, each first identifier in a corresponding iteration is determined based on a trade-off between the relative occurrence of the first identifier in the first true positive signature set in the corresponding iteration and the relative occurrence of the first identifier in the first false positive signature set in the corresponding iteration.

[0162] According to an embodiment, the first true positive signature set in a corresponding iteration is included in the previous first true positive signature set in a previous iteration.

[0163] According to an embodiment, another first set of true positive signatures in another corresponding iteration only partially overlaps with another previous first set of true positive signatures in another previous iteration.

[0164] According to an embodiment, a non-transitory computer-readable medium stores instructions, the instructions being used to obtain an initial first set of true-positive signatures and an initial first set of false-positive signatures to be used in a first iteration period of a first iteration process.

[0165] According to an embodiment, the non-transitory computer-readable medium stores instructions, which are used to obtain an initial signature of a road element and generate the initial first set of true positive signatures and the initial first set of false positive signatures based on the initial signature.

[0166] According to an embodiment, a non-transitory computer-readable medium stores instructions that are used to complete a first iteration when (i) a first true positive signature set and a first false positive signature set in a corresponding iteration are approximately equal to (ii) a first true positive signature set and a first false positive signature set in an iteration subsequent to the corresponding iteration.

[0167] According to an embodiment, a non-transitory computer-readable medium stores instructions used to obtain an initial first true positive signature set and an initial first false positive signature set associated with a first iteration of a first iteration process, wherein a first identifier is associated with the first signature true positive signature subset and the first signature false positive signature subset.

[0168] According to an embodiment, a non-transitory computer-readable medium stores instructions used to (i) generate an initial second true positive signature set associated with a first iteration of a second iteration by removing from an initial first true positive signature set any signatures that appear in the first signature true positive signature subset, and (ii) generate an initial second false positive signature set associated with a first iteration of a second iteration by removing from the initial first false positive signature set any signatures that appear in the first signature false positive signature subset.

[0169] According to an embodiment, the non-transitory computer readable medium stores instructions for generating additional signatures until convergence is reached.

[0170] According to an embodiment, the first signature and the second signature relate to road elements represented by the acquired information.

[0171] According to an embodiment, the first set of false positive signatures represents road elements similar to the road element.

[0172] According to an embodiment, the non-transitory computer-readable medium stores instructions for generating additional signatures to provide a signature cluster associated with the road element and granting an inference process access to the cluster.

[0173] In the foregoing detailed description, numerous specific details are set forth to provide a thorough understanding of the present invention. However, it will be understood by those skilled in the art that the present invention may be practiced without these specific details. In other instances, well-known methods, programs, and components have not been described in detail so as not to obscure the present invention.

[0174] While the concluding portion of the specification particularly points out and distinctly claims the subject matter of the present invention, the organization and method of operation of the invention, together with its objects, features and advantages, can best be understood by reference to the following specific embodiments when read in conjunction with the drawings.

[0175] It will be understood that for simplicity and clarity of illustration, elements shown in the figures have not necessarily been drawn to scale. For example, the size of some elements may be exaggerated relative to other elements for clarity. Furthermore, where appropriate, reference numerals may be repeated among the figures to indicate corresponding or similar elements.

[0176] Because most of the embodiments of the present invention can be implemented using electronic components and circuits known to those skilled in the art, no more detail will be described than is deemed necessary to allow an understanding and appreciation of the basic concepts of the present invention and to avoid obscuring or distracting the teachings of the present invention.

[0177] Any reference in the specification to a method should apply mutatis mutandis to a device or system capable of performing the method and / or to a non-transitory computer-readable medium storing instructions for performing the method.

[0178] Any reference in the specification to a system or device should be applied mutatis mutandis to a method executable on the system and / or may be applied mutatis mutandis to a non-transitory computer-readable medium storing instructions executable on the system.

[0179] Any reference in the specification to a non-transitory computer-readable medium should apply mutatis mutandis to a device or system capable of executing instructions stored on the non-transitory computer-readable medium and / or may apply mutatis mutandis to a method for executing instructions.

[0180] Any combination of any modules or units listed in any one of the drawings, any part of the specification and / or any claim may be provided.

[0181] Any one of the transformation module, active learning module, or clustering module, or any other module described herein, may be performed by hardware and / or code, instructions, and / or commands stored on a non-transitory computer-readable medium, and may be included in the vehicle, outside the vehicle, in a mobile device, in a server, etc.

[0182] The vehicle may be any type of vehicle, for example, a ground vehicle, an air vehicle, or a watercraft.

[0183] The description and / or drawings may relate to images. An image is an example of a media unit. Any reference to an image may apply mutatis mutandis to a media unit. A media unit may be an example of sensory information. Any reference to a media unit may apply mutatis mutandis to any type of natural signal, such as, but not limited to, naturally generated signals, signals indicative of human behavior, signals indicative of stock market-related operations, medical signals, financial sequences, ground measurement signals, geophysical, chemical, molecular, textual and digital signals, time sequences, etc. Any reference to a media unit may apply mutatis mutandis to sensory information. Sensory information may be of any kind and may be sensed by any type of sensor, such as a visible light camera, audio sensor, sensitive infrared, radar imaging, ultrasound, electro-optical, radiographic, laser radar (light detection and ranging), etc. Sensoring may include generating samples (e.g., pixels, audio signals) indicative of a signal transmitted or otherwise reaching the sensor.

[0184] The specification and / or drawings may refer to span elements. The span elements may be implemented as software or hardware. Different span elements of a particular iteration are configured to apply different mathematical functions to the inputs they receive. Non-limiting examples of mathematical functions include filtering, although other functions may also be applied.

[0185] The description and / or drawings may relate to a conceptual structure. The conceptual structure may include one or more clusters. Each cluster may include a signature and correlating metadata. Each citation to one or more clusters may apply to a citation to the conceptual structure.

[0186] The specification and / or drawings may relate to a processor. The processor may be a processing circuit. The processing circuit may be implemented as a central processing unit (CPU) and / or one or more other integrated circuits, such as an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), a custom integrated circuit, or the like, or a combination of such integrated circuits.

[0187] Any combination of any steps of any method illustrated in the specification and / or figures may be provided.

[0188] Any combination of any subject matter of any claim may be provided.

[0189] Any combination of the systems, units, components, processors, sensors shown in the specification and / or figures may be provided.

[0190] Any reference to an object may apply to a pattern, and therefore any reference to object detection applies mutatis mutandis to pattern detection.

[0191] A situation may be a single location / property combination at a point in time. A scene is a sequence of events that logically occur within a causal reference framework. Any reference to a scene applies mutatis mutandis to a situation.

[0192] The sensory information units may be sensed by one or more sensors of one or more types, and the one or more sensors may belong to the same device or system or to different devices or systems.

[0193] A perception unit may be provided, and prior to the perception unit there may be one or more sensors and / or one or more interfaces that receive one or more sensory information units from it. The perception unit may be configured to receive the sensory information units from the I / O interface and / or the sensors. After the perception unit there may be a plurality of narrow AI agents, also referred to as a set of narrow AI agents.

[0194] The sensory information units may or may not be processed before reaching the perception unit. Optional processing - filtering, noise reduction, etc. may be provided.

Claims

1. 1. A computer-implemented method for use in generating a dynamic correlation signature, comprising: generating a first signature in a first iteration, the first signature including a first identifier indicating at least one of (a) a road element characteristic associated with the first signature, or (b) the generated characteristics of the first signature, the first identifiers being generated in association with each other, and determining the first identifier for each iteration of the first iteration based on a relative occurrence of the first identifier with respect to a corresponding first true positive signature set and a relative occurrence of the first identifier with respect to a corresponding first false positive signature set, thereby determining a first true positive signature set and a first false positive signature set for each iteration of the first iteration based on a previous first true positive signature set and a previous first false positive signature set, respectively; generating a second signature in a second iteration, the second signature including (a) a road element characteristic associated with the second signature, or (b) a second identifier indicating the generated characteristic of the second signature, the second identifiers being generated in association with each other, and determining the second identifier for each iteration of the second iteration based on a relative occurrence of the second identifier with respect to a corresponding second true positive signature set and a relative occurrence of the second identifier with respect to a corresponding second false positive signature set, so that for each iteration of the second iteration, a second true positive signature set and a second false positive signature set are determined based on a previous second true positive signature set and a previous second false positive signature set; a second identifier of the second signature is generated independently of a first identifier of the first signature, and the first signature and the second signature collectively represent a sensed information cluster. method.

2. generating a third signature in a third iteration, the third signature including (a) a road element characteristic associated with the third signature, or (b) a third identifier indicating the generated characteristic of the third signature, the third identifiers being generated in association with each other, and determining the third identifier for each iteration of the third iteration based on a relative occurrence of the third identifier with respect to a corresponding third true positive signature set and a relative occurrence of the third identifier with respect to a corresponding third false positive signature set, so that a third true positive signature set and a third false positive signature set are determined for each iteration of the third iteration based on the previous third true positive signature set and the previous third false positive signature set; a third identifier of the third signature is generated independently of a second identifier of the second signature, and the second signature and the third signature collectively represent a sensed information cluster. The method of claim 1.

3. the first signature represents the content of a plurality of sensory information units; The method of claim 1.

4. The method of claim 1 , wherein the first identifier represents non-zero bits of a sparse representation of a neural network feature vector.

5. the first identifier represents an activated neuron of a neural network; The method of claim 1.

6. generating additional signatures until convergence is reached; The method of claim 1.

7. The method of claim 1 , wherein the first signature and the second signature are associated with road elements represented by acquired information.

8. the first set of false positive signatures represents road elements similar to the road element; The method of claim 7.

9. generating additional signatures to provide a signature cluster associated with the road element and granting an inference process access to the cluster; The method of claim 7.

10. determining the first set of true positive signatures and the first set of false positive signatures in association with perception data for one of the iterations of the first iteration; The method of claim 1.

11. determining the first set of true positive signatures and the first set of false positive signatures in association with the sensed information for one of the iterations of the first iteration; The method of claim 1.

12. determining the first set of true positive signatures and the first set of false positive signatures associated with a target object for one of the iterations of the first iteration; The method of claim 1.

13. further comprising analyzing the first generated signature and the second generated signature in a real-time application. The method of claim 1.

14. providing the first generated signature and the second generated signature to an offline application and analyzing the first signature and the second signature in the offline application. The method of claim 1.

15. A non-transitory computer readable medium for use in generating a dynamic correlation signature, the non-transitory computer readable medium storing instructions that, when executed by a processing circuit, cause the processing circuit to perform the method of any one of claims 1 to 14. Non-transitory computer-readable medium.

Images