Efficient management of facial recognition systems and methods in multiple domains

Abstract

The system and method can improve facial recognition processing capabilities by reducing search data as a person traverses a theme park. The system and method include storing facial identification data in a theme park database when a guest enters the theme park and deleting the facial identification data from the theme park database when the guest leaves the theme park. In this manner, the theme park database can be reduced in size as guests leave the theme park because it contains facial identification data for guests within the theme park but not for guests who have left the theme park. Similarly, this approach can be implemented in each of the various areas of the theme park, such that the size of the database associated with each area of ​​the theme park includes only entries corresponding to guests who are in that particular amusement park area.

Description

[Technical Field]

[0001] This disclosure relates generally to facial recognition and, more particularly, to managing the collection and storage of facial recognition identification data. [Background technology]

[0002] This section is intended to introduce the reader to various aspects that may be related to various aspects of the present disclosure, as described below. This discussion is believed to be helpful in providing the reader with background information to facilitate a better understanding of the various aspects of the present disclosure. As such, it should be understood that this description is to be read in this light, and not as admissions of prior art.

[0003] Many venues, such as theme parks and concert venues, manage access rights for large numbers of visitors. Such venues often utilize traditional queuing systems to control access and manage capacity. For example, traditional paper ticket systems are often utilized to verify access rights, obtain visitor information, and / or control the number of visitors admitted to the venue within a specific time frame. More modern technologies for verifying access rights, retrieving visitor information, and / or controlling the number of visitors admitted may include the use of biometric data, such as fingerprint scanning or facial recognition technology. The use of biometric data may require significant data storage and computing resources, increasing the computational complexity associated with verifying access rights and controlling the number of people admitted. As a result, verifying access rights using these modern technologies may result in delayed and / or inaccurate verification (e.g., false positives). The computational complexity associated with these modern technologies increases as the number of visitors increases, for example, during periods of seasonal demand.

[0004] The use of biometric data would be attractive for verifying access rights and controlling the number of people admitted, as visitors would typically not need to carry traditional paper tickets. Summary of the Invention [Problem to be solved by the invention]

[0005] However, existing techniques for efficiently managing large amounts of biometric data need to be improved, and their implementation can be difficult to implement and coordinate across a variety of large-scale venues. [Means for solving the problem]

[0006] Certain embodiments commensurate in scope with the originally claimed subject matter are summarized below. These embodiments are not intended to limit the scope of the present disclosure; rather, these embodiments are intended only to provide a brief summary of certain disclosed embodiments. Indeed, the present disclosure may include a variety of forms that may be similar to or different from the embodiments set forth below.

[0007] The present disclosure provides systems and methods that can improve facial recognition processing capabilities, reduce the time it takes to process facial recognition access / requests, reduce the amount of facial recognition data used for search, and / or improve the accuracy of verifying a guest's facial identity by reducing search data as the guest traverses a theme park. The disclosed systems and methods include storing facial identification data in a theme park database when a guest enters the theme park and deleting the facial identification data in the theme park database when the guest leaves the theme park. In this manner, the theme park database can be reduced in size as guests leave the theme park because it only includes facial identification data for guests within the theme park and not for guests who have left the theme park. Similarly, this approach can be implemented in each of the various areas of a theme park such that the size of the database associated with each area of ​​the theme park only includes entries corresponding to guests within that particular amusement park area.

[0008] In one embodiment, a system for improving facial recognition performance includes an image capture device associated with an area within a venue for capturing facial images. The venue is associated with a first database, and the area is associated with a second database. The system also includes a computer system communicatively coupled to the image capture device. The computer system includes a processing circuit and a memory device communicatively coupled to the processing circuit. The memory device stores executable instructions for causing the processing circuit to perform operations. The operations include receiving a facial image, determining a sufficient correlation between the facial image and a plurality of entries of facial identification data stored in the first database, and adding the facial image to the second database in response to the facial image being sufficiently correlated with the entry.

[0009] In one embodiment, a method is provided that is executed by a processing circuit. The method includes identifying entrance facial identification data from a first facial image captured by an entrance image capture device located at an entrance to a venue. The method includes storing the entrance facial identification data in a venue database that stores a plurality of facial identification data entries. The method includes transferring the entrance facial identification data from the venue database to a region database associated with the venue region in response to a second facial image captured at an entrance to the venue region sufficiently correlating with the entrance facial identification data. The method includes receiving, by the processing circuit, a third facial image captured by an exit image capture device located at an exit of the venue, and determining a sufficient correlation between exit facial identification data associated with the third facial image and an entry in the plurality of facial identification data entries. The method includes deleting the entry in the plurality of entrance facial identification data entries from the venue database in response to determining that the exit facial identification data sufficiently correlates with an entry in the plurality of facial identification data entries.

[0010] In one embodiment, a facial recognition system for improving facial recognition processing capabilities includes a first image capture device configured to capture a first facial image at an entrance to a theme park and a plurality of second image capture devices corresponding to a plurality of areas of the theme park, the plurality of second image capture devices capturing a second facial image at an entrance to a corresponding area of ​​the plurality of areas of the theme park. The facial recognition system further includes a storage device including a theme park database of facial identification data associated with the first facial image and a plurality of amusement park area databases. Each amusement park area database of the plurality of amusement park area databases is associated with a respective area of ​​the plurality of areas, and each amusement park area database of the plurality of amusement park area databases includes facial identification data associated with a respective second facial image. The facial recognition system further includes a computer system communicatively coupled to the first image capture device, the plurality of second image capture devices, and the storage device. The computer system includes a processing circuit and a memory device storing instructions for causing the processing circuit to perform operations. The operations include receiving the first facial image, generating corresponding facial identification data, and storing the corresponding facial identification data in the theme park database. The operations include receiving a particular second facial image captured by a particular second image capture device of the plurality of second image capture devices to determine a sufficient correlation between the particular second facial image and corresponding facial identification data, the particular second image capture device being associated with a particular region of the plurality of regions. Further, the operations include adding the corresponding facial identification data from the theme park database to a particular amusement park region database associated with the particular area based on the sufficient correlation.

[0011] These and other features, aspects, and advantages of the present disclosure will be better understood when the following detailed description is read in conjunction with the accompanying drawings, in which like characters represent like parts throughout the drawings. [Brief explanation of the drawings]

[0012] [Figure 1]FIG. 1 is a schematic diagram of a theme park including multiple amusement park areas using facial recognition technology for visitor management, according to aspects of the present disclosure. [Figure 2] FIG. 2 is a block diagram of a facial recognition system that uses facial recognition technology by using a theme park database and an amusement park area database corresponding to the amusement park area of ​​FIG. 1 according to an aspect of the present disclosure. [Figure 3] FIG. 2 is a flow diagram of a process for improving the facial recognition processing capabilities of the theme park of FIG. 1 according to an aspect of the present disclosure. [Figure 4] 4 is a perspective view of a guest entering the theme park of FIG. 1 and activating the facial recognition system of FIG. 2 to execute the process of FIG. 3, according to an embodiment of the present disclosure. [Figure 5] 3 is a block diagram of the theme park database of FIG. 2 and the amusement park area database of FIG. 2 according to an aspect of the present disclosure. [Figure 6] 4 is a perspective view of a guest leaving the theme park of FIG. 1 and activating the facial recognition system of FIG. 2 to execute the process of FIG. 3, according to an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0013] One or more specific embodiments are described below. In the interest of providing a concise description of these embodiments, not all features of an actual implementation are described herein. It will be appreciated that, as with any industrial design or engineering project, the development of any such actual implementation will require numerous implementation-specific decisions to be made in order to achieve the developers' particular goals, including compliance with system-related and business-related constraints that may vary from implementation to implementation. It will further be appreciated that such development efforts may be complex and time-consuming, but will nevertheless represent a routine undertaking of design, fabrication, and manufacture for those of ordinary skill in the art having the benefit of this disclosure.

[0014] When describing elements of various embodiments of the present disclosure, the articles "a," "an," and "the" are intended to mean that there are one or more of the element. The terms "comprising," "including," and "having" are intended to be inclusive and mean that there may be additional elements other than the listed elements. In addition, it should be understood that references to "one embodiment" or "one embodiment" of the present disclosure are not intended to exclude the existence of additional embodiments that incorporate the referenced features.

[0015] While the following description is generally presented in the context of theme parks, it should be understood that the embodiments disclosed herein are not limited to such entertainment contexts. Indeed, the systems, methods, and concepts disclosed herein can be implemented in a wide variety of applications. Examples are provided in this disclosure to facilitate explanation of the disclosed technology by providing examples of real-world implementations and applications. It should be understood that the embodiments disclosed herein are useful in many applications, such as event venues (e.g., concerts), restaurants, large commercial buildings, hospitals, and / or other industrial, commercial, and / or entertainment systems serving a wide variety of clients / users, to name a few additional examples.

[0016] With this in mind, the present disclosure provides systems and methods for improving facial recognition processing capabilities, reducing processing time for facial recognition access, reducing the amount of facial recognition data used for searching, and / or improving the accuracy of verifying facial identification information by reducing search data as guests traverse a theme park (e.g., an amusement park or entertainment venue) and engage with various functions enabled by facial recognition technology (e.g., accessing theme park rides, paying for food, accessing profile data, etc.). As described above, certain venues, such as theme parks, may use facial recognition technology to verify guests' identities and access rights, obtain information related to guests, and further control the number of people admitted to the venue within a particular time frame. Successful implementation of facial recognition technology requires large amounts of data storage, which can increase the computational complexity associated with verifying access rights and a person's identity. As a result, verifying access rights through these state-of-the-art facial recognition technologies can result in delays and / or inaccurate verification (e.g., false positives).

[0017] For example, determining whether a database of facial identification data stores information corresponding to a guest may include determining that a newly acquired facial image of the guest substantially correlates (e.g., within a target percentage of accuracy) with a facial identification data entry stored in the database. While simplifying the calculation by examining fewer facial features may increase the speed at which guests are identified, this approach may lead to false positives (e.g., the guest may be associated with the wrong facial identification data entry). Thus, a need exists for improving the speed at which a person's identity is confirmed using facial recognition technology without compromising the accuracy of the identification. As used herein, "facial identification data" may refer to identification information extrapolated from a facial image and associated with a particular profile of the guest to facilitate facial matching and match-based access authorization. As used herein, "logic" may refer to hardware, software, or both executable to perform a series of steps. While the following description includes descriptions related to determining a "match" between a captured image and a stored image, it should be understood that a "match" may refer to a sufficient correlation between a captured image and a stored image using the techniques disclosed below.

[0018] The present disclosure includes systems and methods that address this issue by storing facial identification data in a theme park database when a guest enters a theme park and deleting the facial identification data in the theme park database when the guest leaves the theme park. In this manner, the theme park database includes facial identification data for guests within the theme park (but not for guests who have left the theme park), thereby reducing the size of the theme park database when guests leave the theme park. As used herein, a "theme park database" refers to a database (e.g., a Structured Query Language [SQL] database, an Unstructured Query Language [NoSQL] database, etc.) that includes facial identification information for guests entering an entire theme park. As used herein, an "amusement park area database" refers to a database (e.g., an SQL database, a NoSQL database, etc.) that includes facial identification data for guests in a particular area or zone of a theme park. For example, an entry in an amusement park area database may include a smaller subset of the entries in the theme park database that correspond to the amusement park areas of the theme park. Each amusement park area in a theme park may include a corresponding amusement park area database.

[0019] The computer system may receive a facial image of a guest (e.g., captured by an image capture device) as the guest enters the theme park and store corresponding facial identification data. When the guest enters a particular area within the theme park, the computer system may receive another facial image of the guest, match the facial image with facial identification data in a theme park database, and then store the matching facial identification data from the theme park database in an amusement park area database that corresponds to the particular area within the theme park. In this way, as the guest interacts with the facial recognition technology (e.g., image capture device) in a particular area, the facial recognition calculations are performed on the smaller amusement park area database rather than the larger theme park database, thereby improving the accuracy and speed of performance of the facial recognition technology.

[0020] The computer system can receive additional facial images as guests leave a particular amusement park area, causing the computer system to delete the matching facial identification data from the amusement park area database. In this manner, the size of the amusement park area database is reduced as guests leave the amusement park area, thereby reducing the computational burden associated with implementing facial recognition technology in the corresponding amusement park area. The computer system can receive additional facial images as guests leave the theme park, causing the computer system to delete the corresponding facial identification data from the theme park database (and all amusement park area databases). In this manner, the size of the theme park database is reduced as guests leave the theme park, thereby reducing the computational burden associated with implementing facial recognition technology in the theme park.

[0021] While the description of reducing the size of the database has been given in connection with dividing a larger database into smaller databases based on the guest's location (e.g., as captured by an image capture device and verified by a computer system), it should be understood that the size of the database can be reduced based on additional or alternative characteristics of the guest. For example, a larger database can be broken down into respective databases based on whether the guest is a regular theme park attendant or whether the guest is an infrequent theme park attendant (e.g., a park hopper). As another example, a database can be divided into smaller databases, such as a smaller database for guests with wearable devices and another smaller database for guests without wearable devices. Furthermore, wearable devices can be of a first type or a second type, and thus the smaller database can be further divided into an even smaller database associated with guests wearing wearable devices of the first type and another even smaller database associated with guests wearing wearable devices of the second type.

[0022] With the above in mind, FIG. 1 is a schematic diagram of a theme park 100, including an amusement area, that utilizes facial recognition technology for visitor management in accordance with an aspect of the present disclosure. Theme park 100 includes a computer system 102, an image capture device 104, and a communications network 106, as well as other components that cooperate in accordance with the present disclosure as described in detail below. Certain aspects of theme park 100 will be referenced with respect to the steps they perform or facilitate, as illustrated in the process depicted in FIG. 3. Specifically, it is noted that the present embodiment facilitates the use of facial recognition technology in theme parks, improving facial recognition processing capabilities without compromising the accuracy of the facial recognition technology. In this manner, guests 120 may traverse the theme park with little or no physical identification items (e.g., traditional tickets or identification bracelets) that grant them access.

[0023] When guest 120 arrives at an amusement park entrance location 124 (e.g., a hotel front desk, kiosk, guest service counter, amusement park gate), image capture device 104 can capture a facial image of guest 120 and transmit the captured facial image to computer system 102, which can identify and store corresponding facial identification data associated with guest 120 in a theme park database, as discussed in FIG. 3 . In one embodiment, guest 120 can be equipped with additional admission credentials (e.g., a ticket or active wearable device), among other items such as an amusement park map, dining and lodging instructions, and amenity instructions. Information media (e.g., audio, video) and instructional signs indicating instructions for capturing a facial image by image capture device 104 can be present at all such amusement park entrance locations 124. In some implementations, guest 120 can present a facial image to theme park 100 (e.g., by uploading a facial image in advance via the theme park's website) and bypass capturing such credentials at amusement park entrance location 124.

[0024] Guest 120 may enter and gain admission to theme park aisle 126. Specifically, as shown in FIG. 1 , aisle 126 may be devoid of a physical turnstile or similar physical traverse count or control feature that may prevent traversal of aisle 126 when aisle 126 is intended to be open. In one embodiment, after computer system 102 receives the facial image captured by image capture device 104, the computer system activates a physical turnstile to allow guest 120 access to theme park 100. It should be understood that in one embodiment, aisle 126 may be devoid of a turnstile when activated.

[0025] Theme park 100 may use other entry technologies in addition to or in lieu of facial recognition technology for entry. For example, guest 120 may transfer their entry credentials or identification code or number to an active wearable device that has unique identifying information. In this manner, guest 120 may have essentially direct access to the amusement park or amusement park area. For example, guest 120 wearing a properly activated active wearable device may walk through walkway 126 without stopping. Credential information associated with the active wearable device held or worn by guest 120 may be provided to gate monitoring equipment 128, where it may be detected and monitored by amusement park employees (e.g., a security services company). Furthermore, activated active wearable devices may be associated with the same guest profile associated with captured facial images stored by computer system 102. Guest 120 may decline to use the theme park's facial recognition technology and instead accept the use of another access authorization method, such as a wearable device. Accordingly, the embodiments disclosed herein may be used by guests utilizing wearable devices, facial recognition technology, and / or other methods for authorizing and / or tracking guest access.

[0026] Additionally, although the present disclosure is described in the context of facial images being used for identification and access control purposes, other biometric data such as fingerprints or other modes of identification (e.g., eye recognition) can be used. Additionally, the computer system can receive information from the guest 120 by any suitable device, such as a pocketable wireless device, a wearable device, a handheld device, or a mobile device, via a transmitter / receiver located on the device. In one embodiment, the guest-related device can be waterproof.

[0027] After the computer system 102 stores the facial identification data of the guest 120, the guest 120 is granted access to the theme park 100. The computer system 102 can associate a profile for each guest 120 with the guest's corresponding facial identification data. For example, the guest profile can include access to certain features of the theme park and limit access to other features of the theme park 100. In this manner, the computer system can grant the guest 120 access to authorized features of the theme park 100 corresponding to the permissions defined by the guest profile. As used herein, such authorization information and personal information associated with the guest can be referred to as "profile data." Access to the theme park 102 can be granted in response to the computer system 102 operating a turnstile, lifting a gate, unlocking a door, and / or any other suitable actuatable device. In the absence of any actuatable device, the computer system 102 can alert security personnel in response to receiving an indication of the guest being in a restricted area.

[0028] As discussed above, each guest profile can alternatively or additionally be broken down based on data other than facial identification data. The database can be divided into smaller databases, such as a smaller database for guests with wearable devices and another smaller database for guests without wearable devices. Furthermore, the wearable devices can be of a first type or a second type, such that the smaller database can be further divided into a smaller database associated with guests wearing wearable devices of the first type and another smaller database associated with guests wearing wearable devices of the second type.

[0029] As illustrated, theme park 100 includes various amusement park areas 130. For example, theme park 100 may include a first amusement park area 132 that includes a water park 134, a second amusement park area 140 that includes a restaurant area 142, and a third amusement park area 150 that includes a ride system 152. Although the illustrated theme park 100 includes three amusement park areas 130, it should be understood that theme park 100 may include any suitable number of amusement park areas 130 having any suitable characteristics.

[0030] The amusement park area 130 may include an amusement park entrance 160 that includes a corresponding image capture device 104. In the illustrated schematic diagram, the amusement park area 130 is defined and bounded by a dashed line. As described in more detail below, when a guest 120 enters the amusement park area 130 through the amusement park entrance 160, the image capture device 104 may capture a facial image of the guest 120 and transmit it to the computer system 102 via the communications network 106 (e.g., for verification / authentication purposes). For example, in association with a first amusement park area 132, the computer system 102 may receive a live facial image of the guest 120 seeking access to the first amusement park area 132 and determine whether the facial image matches facial identification data of an existing entry in the theme park database. In response to the facial image matching the facial identification data of an existing entry in the theme park database, the computer system 102 may copy the matching facial identification from the theme park database to the park database associated with the first amusement park area 132.

[0031] If a match is not determined, computer system 102 may add the facial image to a list of "missing" images, track guests using cameras within the amusement park, attempt to acquire subsequent facial images to match the facial image to facial identification data of existing entries in the theme park database, and / or take any other appropriate action. In one embodiment, in response to determining that the facial image does not match an existing entry in the theme park database, computer system 102 may alert security personnel.

[0032] Additionally, features of amusement park region 130 may also include image capture devices 104. For example, water park 134 may include image capture devices 104 to determine whether guest 120 has purchased an access package granting access to water park 134. To make this determination, computer system 102 may receive facial images captured by image capture device 104 at water park 134 and determine whether the captured facial images match facial identification data (in the region database) corresponding to entries with access rights to water park 134.

[0033] In another example, restaurant area 142 may include image capture devices 104 to facilitate payment for food. For example, computer system 102 may receive live facial images captured by image capture devices 104 in restaurant area 142, match the facial images to facial identification data (in the area database), and charge a deposit account associated with the matching facial identification data. This may be particularly advantageous for guests 120 who wish not to use cards, cash, or other payment methods within theme park 100 (e.g., because guest 120 may be exposed to water).

[0034] In yet another example, the ride system 152 may include an image capture device 104 to determine whether a guest 120 may access the ride system 152. To make this determination, the computer system 102 may receive facial images captured by the ride system's image capture device 104 and determine whether the live facial image matches facial identification data (in the area database) corresponding to an entry with access to the ride system 152. As discussed above, it should be understood that a particular guest 120 may decline use of the theme park's facial recognition technology and instead accept the use of another access authorization method, such as a wearable device. Accordingly, the determination of whether to grant access to the ride system 152 may alternatively be based on access rights associated with a wearable device having unique identification information corresponding to the guest's profile data. Guests may also be tracked using the wearable device. Additionally, the determination of whether to grant access may be based on the guest's age, weight, or height, such that access is restricted to guests 120 who do not meet age, weight, or height requirements.

[0035] 2 is a block diagram of a facial recognition system 200 that employs facial recognition technology by using a theme park database and an amusement park area database corresponding to the amusement park areas of FIG. 1 , according to an embodiment of the present disclosure. Facial recognition system 200 may include one or more image capture devices 104. Image capture device 104 may include a camera 202 capable of capturing facial images 203 of guest 120 ( FIG. 1 ). For example, image capture device 104 may include any suitable device that may include camera 202, such as a digital camera, an action camera, a high-speed camera, a hidden camera, a webcam connected to a tablet computer, a desktop or laptop, and / or any suitable device capable of capturing facial images of guest 120.

[0036] Image capture device 104 may include a display 204 capable of presenting instructions to guest 120. Display 204 may include a liquid crystal display (LCD), a segmented display, a light emitting diode (LED) display, an organic light emitting diode (OLED) display, and / or any suitable device capable of displaying information. For example, display 204 may display a countdown to guest 120 to capture an image, a message that an attempt to capture an image failed, a message indicating that the image needs to be captured again (e.g., because of insufficient quality), a message that the image was successfully captured, and / or any other suitable content to guest 120 or park personnel.

[0037] Image capture device 104 may also include a controller 210 that controls image capture device 104. Controller 210 may include a processor 212 that executes instructions for controlling image capture device 104 and a memory 214 that stores instructions. Image capture device 104 may include a communications interface 216 that enables image capture device 104 to communicate with other devices, such as computer system 102, over communications network 106. More information regarding processor 212, memory 214, communications interface 216, and communications network 106 is provided below. It should be understood that theme park 100 (FIG. 1) may include any suitable number of image capture devices 104.

[0038] Computer system 102 may include a storage device 220, a communication interface 236, and a controller 230. Controller 230 may include a processor 232, a memory 234, and a communication interface 236. More details regarding storage device 220, processor 232, memory 234, and communication interface 236 are provided below.

[0039] The computer system 102 may execute the face identification logic 240 to perform the process described below with respect to Figure 3. The face identification logic 240 may be stored as machine-readable instructions stored in the memory 234 and executed by the processor 232. In one embodiment, the face identification logic 240 may be adjusted or tuned to set conditions for acceptable live images captured by the image capture device 104 and to set criteria for suitable and unsuitable live images.

[0040] For example, the facial identification logic 240 may include artificial intelligence (AI) logic that may track successful and / or unsuccessful attempts to acquire facial images having desired characteristics (e.g., sufficient to distinguish one guest from another or sufficient representation of the facial geometric features of the guest 120) based on the clarity of the acquired facial images. The AI ​​logic may adjust or adjust parameters for determining acceptable facial images acquired by the image capture device 104 to improve success and efficiency over time. In one embodiment, the AI ​​logic may define thresholds (e.g., percentages, ratios, etc.) for the image quality of the acquired facial images of the guest 120 (e.g., based on the percentages, ratios, etc. of desired characteristics in the acquired facial images). For example, if the thresholds for the parameters defining the image quality of the acquired facial images are insufficient, the artificial intelligence logic may increase or decrease the respective thresholds to improve or optimize guest throughput within the theme park 100.

[0041] As described above, each of the image capture device 104 and the computer system 102 may include a respective controller 210, 230 including one or more respective processors 212, 232 and one or more respective memory devices 214, 234. The processors 212, 232 (e.g., microprocessors) may execute software programs and / or instructions that facilitate capturing and analyzing the facial image 203, determining whether the captured facial image 203 matches existing facial identification data 242, storing or deleting the facial identification data 242 from the storage device 220, etc. Additionally, the processors 212, 232 may include multiple microprocessors, one or more “general-purpose” microprocessors, one or more special-purpose microprocessors, and / or one or more application-specific integrated circuits (ASICs), or some combination thereof. For example, the processors 212, 232 may include one or more reduced instruction set computer (RISC) processors.

[0042] The memory devices 214, 234 and the storage device 220 may store information such as control software, look-up tables, configuration data, etc. In one embodiment, the processors 212, 232, the memory devices 214, 234, and / or the storage device 220 may be external to the respective controllers 210, 230. Each of the memory devices 214, 234 and the storage device 220 may include a tangible, non-transitory, machine-readable medium such as volatile memory (e.g., random access memory (RAM)) and / or non-volatile memory (e.g., read-only memory (ROM)). Each of the memory devices 214, 234 and the storage device 220 may include one or more storage devices (e.g., non-volatile storage devices), which may include read-only memory (ROM), flash memory, a hard drive, and / or any other suitable optical, magnetic, or solid-state storage medium.

[0043] Memory devices 214, 234 and storage device 220 may store various information used for various purposes. For example, memory devices 214, 234 and storage device 220 may store machine-readable and / or processor-executable instructions (e.g., firmware or software) for execution by processors 212, 232, such as instructions that facilitate the acquisition and / or storage of facial identification data 242 associated with facial images acquired by image capture device 104.

[0044] The computer system 102 may receive the facial image 203 and store the facial image 203 and corresponding facial identification data 242 in the storage device 220. As described in more detail below, the storage device 220 may include a theme park database 250, an amusement park area database 252, and profile data 254. The profile data 254 may include information about each of the guests 120. For example, the profile data 254 may include identification information (e.g., name, age, height, weight, date of birth, etc.), access information (e.g., packages purchased by a particular guest 120, packages granting a particular guest 120 specific access / restrictions, etc.), etc. The profile data 254 may be stored in a database, e.g., a database having information about each guest 120 in a respective table along with the profile data 254 in a corresponding entry in the table. In one embodiment, the computer system 102 can associate the facial identification data 242 with the profile data 254, and using the facial recognition techniques described herein causes the computer system 102 to reference the profile data 254 based on the facial image 203.

[0045] Computer system 102 can store facial identification data 242 in theme park database 250 when a corresponding guest 120 enters theme park 100 and can delete facial identification data 242 from theme park database 250 when the corresponding guest 120 exits theme park 100. In this manner, theme park database 250 only includes facial identification data 242 for guests 120 within theme park 100, thereby reducing the size of theme park database 250 as guests exit theme park 100.

[0046] Additionally, when guest 120 enters a particular area 130 (FIG. 1) within theme park 100, computer system 102 may receive a new facial image 203 of guest 120 from image capture device 104 associated with the particular area 130. Computer system 102 may match the new facial image 203 with facial identification data 242 in theme park database 250 and then store the facial identification data 242 from theme park database 250 in amusement park area database 252 corresponding to the particular area 130. In this manner, when guest 120 interacts with image capture device 104 in the particular area 130, facial recognition calculations are performed in the smaller amusement park area database 252 rather than the larger theme park database 250, thereby improving the accuracy and speed of performance of the facial recognition techniques described above.

[0047] Additionally, the image capture device 104, the computer system 102, and other features of the theme park 100 may include respective communication interfaces 216, 236 communicatively coupled to the respective controllers 210, 230, which may communicate with any suitable communication network, including the communication network 106. For example, the communication network 106 may include a wireless network, such as a mobile network, Wi-Fi, Bluetooth, a local area network (LAN), a wide area network (WAN), the Internet, or the like. The communication interfaces 216, 236 may enable the respective controllers 210, 230 to communicate with the communication network 106. In this manner, the communication interfaces 216, 236 may enable the controllers 210, 230 to communicate with each other to enable the exchange of information to facilitate the implementation of facial recognition technology.

[0048] FIG. 3 is a flow diagram of a process 300 for improving the facial recognition processing capabilities of the theme park 100 of FIG. 1 in accordance with aspects of the present disclosure. In one embodiment, the process 300 may be performed by the facial recognition system 200 of FIG. 2. To facilitate the description of the process 300, FIGS. 3 through 6 are described together. That is, when describing the process 300, the present disclosure may refer to FIGS. 4 through 6. FIG. 4 is a perspective view of a guest 120 (FIG. 1) entering the theme park 100 of FIG. 1 in accordance with aspects of the present disclosure. FIG. 5 is a block diagram of the theme park database 250 of FIG. 2 and the amusement park area database 252 of FIG. 2 in accordance with aspects of the present disclosure.

[0049] Process 300 may be performed by facial recognition system 200 of FIG. 2 to perform facial recognition techniques in a more efficient manner. In one embodiment, computer system 102 may perform process 300. For example, process 300 may be in the form of non-transitory computer-readable instructions executable by suitable processing circuitry (e.g., processor 232 of FIG. 2) of computer system 102. At least some of the steps of process 300 may be performed by using AI and / or machine learning techniques. Process 300 is provided by way of example, and certain illustrated steps of process 300 may be performed in other orders, skipped, or repeated in accordance with the present disclosure.

[0050] Process 300 includes receiving a facial image 203 (FIG. 2) from an image capture device 104 (FIGS. 1-2) at an amusement park entrance location 124 (FIG. 1) (process block 310). Referring to FIG. 4, after waiting in the amusement park entrance queue, image capture device 104 at amusement park entrance location 124 (FIG. 1) can capture a facial image 203 of guest 120 via camera 202. Image capture device 104 (or computer system 102) can determine whether facial image 203 is of sufficient quality, and target facial features can be identified with sufficient precision to determine guest 120's facial geometry or a threshold number of facial features that are identifiable or distinguishable from other guests. Image capture device 104 can cause camera 202 to attempt to capture a threshold number of facial images 203 or footage of guest 120 moving their face (e.g., rotating, translating, etc.). Additionally or alternatively, the image capture device 104 can capture various facial images 203, and the image capture device 104 (or computer system 102) can select facial images that exceed a quality threshold value or select the top number (e.g., 3, 4, 10, or any preferred number) of facial images 203. In one embodiment, the image quality can be based on an image quality assessment (IQA) that considers image factors such as contrast, blur, noise, artifacts, distortion, or some combination thereof, of the facial images 203, to name a few. However, it should be understood that any suitable IQA technique or techniques can be used to determine the appropriate quality of the facial images 203.

[0051] The computer system 102 can identify facial identification data 242 from the facial image 203 that meets the IQA (e.g., by performing AI or machine learning techniques). For example, the computer system 102 can receive the facial image 203 and, based on the facial image 203, identify the precise locations of certain facial features relative to one another. In one embodiment, the computer system 102 can identify the locations of facial features such as eyebrows, eyes, nose, mouth, facial contours, jawline, ears, hair, and facial hair. The computer system can generate a three-dimensional (3D) or two-dimensional (2D) rendering of the face of the guest 120 ( FIG. 1 ) based on the facial image 203. In addition to or instead of geometry-based face recognition techniques, any suitable face recognition method may use piecemeal-based methods, statistical-based methods (e.g., pixel template matching, statistically-based, principal component analysis (PCA), kernel PCA, discrete cosine transform (DCT), linear discriminant analysis (LDA), locality-preserving projection (LPP), Gabor wavelets, etc.), and / or neural network methods (e.g., neural networks with Gabor filters, neural networks and hidden Markov models (HMMs), fuzzy neural networks), to name a few.

[0052] The guest 120 may have a tangible item, such as a wearable device 312, a mobile device, or an admission ticket, that may contain profile data 254. The computer system 102 may receive information indicative of the profile data 254 and then link the profile data 254 to the facial identification data 242 associated with the captured facial image 203. For example, the guest 120 may pre-purchase VIP access to the theme park 100. The guest 120 may scan a barcode from a mobile device, a beacon from their wearable device 312, or a barcode from their admission ticket at a scanning station 314.

[0053] If an admission ticket is used, the admission ticket may include any suitable document that can be scanned, such as any suitable ticket (e.g., admission ticket, theme park ticket, entertainment ticket, special status pass, etc.), identification document (e.g., passport, driver's license, etc.), form of payment (e.g., credit card, debit card, etc.). Specifically, the admission ticket may provide the guest 120 with a right or entitlement that enables certain privileges (e.g., admission to a theme park or sporting event, access to financial assets, etc.). As such, the guest 120 may scan the admission ticket using the scanning station 314 and transmit information related to the admission ticket to the computer system 102. The computer system 102 may then link and associate the profile data (associated with the admission ticket) with the facial identification data 242 to obtain admission to a theme park, entertainment venue, concert, movie theater, play, special level of the event (e.g., club level or VIP level), etc. Additionally, the admission ticket may serve as an additional or backup method for authenticating the guest 120's identity or for paying for a product or service.

[0054] In either case, the scanning station 314 can read an identification code on the wearable device 312, the mobile device, and / or the admission ticket to authenticate the guest 120. For example, the scanning station 314 can include any suitable type of identification code reader, such as a barcode reader, a quick response (QR) code reader, a radio frequency identification (RFID) code reader, a near field communication tag reader, etc. As such, the identification code can include any suitable type of identification code, such as a barcode, a quick response (QR) code, a radio frequency identification (RFID) code, a near field communication tag, etc. Furthermore, in one embodiment, the admission ticket can be an electronic document and can be displayed on the guest's electronic device, such as a mobile phone (e.g., a smartphone), a tablet computer, the wearable device 312, or any other suitable device capable of displaying an identification code.

[0055] The identification code on the wearable device 312, mobile device, and / or admission ticket can be aligned with the scanning area of ​​the scanning station 314 so that the scanning station 314 can read the identification code. In one embodiment, where the identification code is a barcode or QR code, the scanning station 314 can read elements of the identification code (e.g., markings) by illuminating the identification code with an illumination system (e.g., red light). A sensor in the scanning station 314 can detect the reflected light from the illumination system and generate an analog signal that is sent to a decoder. The decoder can interpret the analog signal, authenticate the identification code (e.g., using a check digit process), and convert the identification code into usable data (e.g., text).

[0056] The scanning station 314 may present the guest 120 via a display 316 with a list of instructions 317 for scanning a tangible item and linking the profile data 254 to the facial identification data 242. Alternatively, the tangible item (e.g., the wearable device 312, the mobile device, and / or the admission ticket) may be omitted, and the guest 120 may engage a user interface 318 (e.g., a button and / or touchscreen interface) to log in to a password-protected profile that includes the profile data 254 and associate the profile data 254 with the facial identification data 242. In this manner, multiple entrance locations 124 may include multiple image capture devices 104 and scanning stations 314, thereby automating and / or speeding the entrance process for the theme park 100 and thereby increasing guest 120 throughput.

[0057] Returning to Figure 3, process 300 includes storing facial identification data in theme park database 250 (process block 320). As mentioned above, theme park database 250 (Figure 2) may include a large database that stores facial identification data 242 of guests 120 in theme park database 250. Because theme park 100 may have the capacity to accommodate hundreds or thousands of guests, performing facial recognition technology on guests 120 within theme park 100 may include matching facial images 203 of guests 120 against facial identification data 242 in theme park database 250. Such an implementation may be computationally intensive for a computer system and may introduce delays.

[0058] To improve facial recognition processing capabilities, theme park 100 may include image capture devices 104 at the entrances to amusement areas 130 ( FIG. 1 ). Computer system 102 may receive facial images 203 via cameras 202 of image capture devices 104 in particular amusement areas 130 (process block 330). Computer system 102 may match facial images 203 captured by image capture devices 104 associated with particular amusement areas 130 against facial identification data 242 in theme park database 250. Matching facial images 203 to facial identification data 242 may include implementing any suitable facial recognition matching technique, such as an EigenFaces facial recognizer, a FischerFaces recognizer, and / or local binary pattern histograms, to name a few.

[0059] In response to a match between facial image 203 captured by image capture device 104 associated with a particular amusement park area 130 and facial identification data 242 of an existing entry in theme park database 250, computer system 102 may copy and store the matching facial identification data 242 from theme park database 250 to the amusement park area database 252 associated with the first amusement park area 132 (process block 340). In one embodiment, storing the matching facial identification data 242 from theme park database 250 to the amusement park area database 252 associated with the first amusement park area 132 (process block 340) may include deleting the matching facial identification data 242 from theme park database 250. In this manner, as guests 120 enter a particular amusement park area 130 within theme park 100, the overall size of theme park database 250 may be reduced to improve throughput of guests 120 entering theme park 100 or various areas 130.

[0060] 5, storage device 220 may maintain one or more repositories of facial identification data 242. Illustratively, storage device 220 may include a theme park database 250 containing facial identification data 242 of guests 120 within theme park 100. Storage device 220 may also include amusement park area databases 252 for each of amusement park areas 130. In this example, storage device 220 includes a first amusement park area database 252A, a second amusement park area database 252B, and an Nth amusement park area database 252C corresponding to n amusement park areas 130.

[0061] In addition to storing facial identification data 242, theme park database 250 may store profile data 254 associated with corresponding guests 120. In the illustrated example, theme park database 250 stores facial identification data 242 and corresponding profile data 254 for ten guests 120, a first amusement park area database 252A stores facial identification data 242 and corresponding profile data 254 for three of the ten guests, and a second amusement park area database 252B stores facial identification data 242 and corresponding profile data 254 for the other three of the ten guests. As described above, when a guest 120 engages with image capture device 104 at an entrance location 124 of theme park 100 and is granted admission to theme park 100, that guest's facial identification data 242 may be newly added and stored in theme park database 250 (process block 320). When a guest 120 moves into a particular theme park area 130, the computer system 102 may copy and store (process block 340) the corresponding facial identification data 242 associated with the guest 120 in the theme park area database 252 associated with the particular theme park area 130. In doing so, the computer system 102 may delete the corresponding facial identification data 242 associated with the guest 120 from the theme park database 250.

[0062] In the illustrated example, theme park database 250 and amusement park area database 252 are shown as storing face identification data 242 and profile data 254, however, it should be understood that in one embodiment, theme park database 250 and / or amusement park area database 252 may include only face identification data 242. In this case, the face identification data may include metadata, pointers, etc., that map corresponding entries in face identification data 242 to corresponding profile data 254. Thus, face identification data 242 and profile data may be located in different databases or in different tables in the same database.

[0063] While the theme park 100 is described as being divided into multiple areas 130, each with a corresponding database, it should be understood that the theme park 100 can be divided into several amusement park areas 130, and the amusement park areas 130 can be divided into several subareas, each with a corresponding database. For example, the theme park 100 can be divided into amusement park areas 130, which can be divided into subareas, which can be divided into amusement rides / buildings, with the theme park 100, areas 130, subareas, and amusement rides / buildings including corresponding image capture devices 104 and databases. Using aspects of the present disclosure in such theme parks can increase facial recognition accuracy and throughput by reducing the database size, and therefore the computational complexity, associated with implementing facial recognition technology.

[0064] 3 , when a guest 120 leaves a particular amusement park area 130 and enters the general theme park 100, an exit image capture device 104 may receive a facial image 203. The computer system 102 may receive the facial image 203 captured at the area exit and delete the facial identification data 242 from the amusement park area database 252 (process block 350). As described above, the computer system 102 may delete from the amusement park area database 252 the facial identification data 242 that most closely matches the facial image 203 captured at the area exit. In this manner, the entries in the amusement park area database 252 may be kept up to date, without entries that are likely to be unused (e.g., entries corresponding to guests 120 who have left the amusement park area 130).

[0065] 5 , deleting face identification data 242 from amusement park area database 252 can cause the amusement park area database to reduce in size. In one embodiment, deleting face identification data 242 from amusement park area database 252 can cause computer system 102 to add the deleted face identification data 242 to theme park database 250. For example, if example guest 3 were to leave first amusement park area 132, computer system 102 could delete face identification data 242 associated with guest 3 from amusement park area database 252 and cause computer system 102 to add face identification data 242 to theme park database 250.

[0066] 6, as guest 120 leaves theme park 100, image capture device 104 at theme park exit 364 may capture facial image 203 of guest 120. Referring to FIG. 3, computer system 102 may receive facial image 203 captured at theme park exit 370 and delete corresponding facial identification data 242 from theme park database 250 (process block 360). Computer system 102 may delete facial identification data 242 from theme park database 250 that best matches facial image 203 captured at theme park exit 364. In this manner, entries in theme park database 250 may be kept up-to-date without entries that are likely to be unused (e.g., entries corresponding to guests 120 who have left theme park 100).

[0067] While only certain features of the disclosed embodiments have been illustrated and described herein, many modifications and changes will occur to those skilled in the art, and it is, therefore, to be understood that the appended claims are intended to cover all such modifications and changes as fall within the true spirit of the present disclosure.

[0068] The technical effects of the present disclosure include systems and methods for improving facial recognition processing capacity, shortening the time to process facial recognition access, and / or improving facial identification verification accuracy as a person moves through a theme park by reducing search data. The technical effects of the present disclosure include storing facial identification data in a theme park database when a guest enters the theme park and deleting the facial identification data from the theme park database when the guest leaves the theme park. In this manner, the theme park database may be reduced in size as guests leave the theme park because it contains only facial identification data for guests within the theme park and not for guests who have left the theme park. Similarly, this approach may be implemented within various areas (and sub-areas) of a theme park, as described above, such that the size of each database associated with a theme park area only contains entries corresponding to guests in that particular amusement park area.

[0069] This specification uses examples of the presently disclosed embodiments, including the best mode, and will enable those skilled in the art to practice the disclosed embodiments, including making and using any devices or systems, and performing any incorporated methods. The patentable scope of the disclosed embodiments is defined by the claims, and may include other examples that occur to those skilled in the art. Such other examples are intended to be within the scope of the claims if they have structural elements that do not differ from the claim language, or if they include equivalent structural elements that do not differ substantially from the claim language.

[0070] The technology shown and claimed herein refers to and applies to tangible objects and specific examples of a practical nature that will materially improve the art, and thus are not abstract, intangible, or purely theoretical. Furthermore, where any claim appended at the end of this specification contains one or more elements designated as "means for [performing] ... [function]" or "step for [performing] ... [function]," such elements are to be construed pursuant to 35 U.S.C. 112(f). Conversely, for any claim containing elements designated in any other manner, such elements are not to be construed pursuant to 35 U.S.C. 112(f). [Explanation of symbols]

[0071] 102 Computer Systems 104 Image acquisition device 106 Communication Network 202 Camera 203 Facial Images 204 Display 210 Controller 212 processors 214 memory 216 Communication Interface 220 Storage Devices 230 Controller 232 processors 234 memory 236 Communication Interface 240 Face ID Logic 242 Face ID Data 250 Theme Park Database 252 Amusement Park Area Database 254 Profile Data

Claims

1. 1. A system for improving face recognition processing capabilities, comprising: a first image capture device configured to capture first facial image data, the first image capture device being associated with an entrance to a venue, the first facial image data being stored in a first database; a second image capture device configured to capture second facial image data, the second image capture device associated with an entrance to an area within the venue, the area associated with a second database; a third image capture device configured to capture third facial image data, the third image capture device associated with an exit of the area of ​​the venue; and a computer system communicatively coupled to the first image acquisition device, the second image acquisition device, and the third image acquisition device; Equipped with The computer system comprises a processing circuit and a memory device communicatively coupled to the processing circuit, the memory device providing to the processing circuit: receiving the second facial image data; determining a first correlation between the second facial image data and the first facial image data; adding the first facial image data to the second database in response to the second facial image data correlating with the first facial image data; receiving the third facial image data; determining a second correlation between the third facial image data and the first facial image data; deleting the first facial image data from the second database in response to the third facial image data correlating with the first facial image data; A system configured to store executable instructions to cause a

2. The instructions cause the processing circuit to: determining the first correlation by comparing first geometric facial features of the first facial image data stored in the first database with second geometric facial features of the second facial image data; The system of claim 1 , wherein the system is executable to:

3. The system of claim 1 , comprising a wearable device communicatively coupled to the computer system.

4. The computer system is configured to receive an identification of the wearable device, and the instructions direct the processing circuitry to: receiving profile data associated with the identification information; linking the profile data to the first facial image data of the guest wearing the wearable device; The system of claim 3 , wherein the system is executable to:

5. The system of claim 4 , wherein the profile data includes identification information for the guest, access information indicating access rights to features of the venue, or both.

6. The instructions cause the processing circuitry to: deleting the first facial image data from the first database in response to the second facial image data correlating with the first facial image data; The system of claim 1 , wherein the system is executable to:

7. The instructions cause the processing circuitry to: adding the first facial image data to the first database in response to the third facial image data correlating with the first facial image data; The system of claim 6 , wherein the system is executable to:

8. The system of claim 1 , wherein the first database is separate from the second database.

9. 1. A method comprising: determining, by a processing circuit, entrance facial identification data from a first facial image captured by an entrance image capture device located at an entrance of the venue; storing, by the processing circuitry, the entrance facial identification data in a venue database, the venue database storing a plurality of facial identification data entries; and in response to the processing circuit correlating a second facial image captured at an entrance to an area of ​​the venue with the entrance facial identification data, transferring the entrance facial identification data from the venue database to an area database associated with the area of ​​the venue. transferring, by the processing circuitry, the entrance facial identification data from the region database to the venue database in response to a third facial image captured at an exit from the region of the venue correlating with the entrance facial identification data; receiving, by the processing circuitry, a fourth facial image captured by an exit image capture device located at an exit of the venue; determining, by the processing circuitry, a correlation between exit facial identification data associated with the fourth facial image and at least one entry of the plurality of facial identification data entries; in response to determining, by the processing circuitry, that the exit facial identification data correlates to one entry of the plurality of facial identification data entries, deleting the determined correlated entry from the venue database; A method comprising:

10. receiving, by the processing circuitry, a fifth facial image from an image capture device associated with the venue feature, the fifth facial image including a live facial image of a guest seeking access to the feature; authenticating the fifth facial image to grant access to the features in response to determining, by the processing circuitry, that the fifth facial image correlates to at least one entry of the plurality of facial identification data entries; 10. The method of claim 9, comprising:

11. receiving a request for guest authentication within the venue; authenticating the guest based on the plurality of facial identification data entries; 10. The method of claim 9, comprising:

12. 10. The method of claim 9, wherein storing the entry facial identification data comprises storing the entry facial identification data in response to the first facial image satisfying one or more image quality parameters.

13. 10. The method of claim 9, including associating, by the processing circuitry, the entrance facial identification data with profile data associated with the guest having the entrance facial identification data.

14. A face recognition system for improving face recognition processing capabilities, comprising: a first image capture device configured to capture a first facial image at an entrance to the theme park; a plurality of second image capture devices corresponding to a plurality of areas of the theme park, each second image capture device of the plurality of second image capture devices configured to capture a second facial image in a corresponding area of ​​the plurality of areas of the theme park; A storage device; a computer system communicatively coupled to the first image acquisition device, the plurality of second image acquisition devices, and the storage device; Equipped with The storage device a theme park database containing facial identification data associated with the first facial image; a plurality of amusement park area databases, each amusement park area database of the plurality of amusement park area databases associated with a respective area of ​​the plurality of areas, each amusement park area database of the plurality of amusement park area databases including face identification data associated with a respective second facial image; Equipped with The computer system includes a processing circuit and a memory device communicatively coupled to the processing circuit, the memory device providing to the processing circuit: receiving the first facial image and generating corresponding facial identification data; and storing the corresponding facial identification data in the theme park database; receiving a first second facial image captured by a first second image capture device of the plurality of second image capture devices; determining a first correlation between the first second facial image and the corresponding facial identification data, the first second image capture device being associated with an entrance to a region of the plurality of regions; adding the corresponding facial identification data from the theme park database to an amusement park area database associated with an area corresponding to the first second image capture device based on the first correlation; receiving a second second facial image captured by a second second image capture device of the plurality of second image capture devices; and determining a second correlation between the second second facial image and the corresponding facial identification data, the second second image capture device being associated with an exit of the region of the plurality of regions; removing the corresponding face identification data from the region database associated with the region based on the second correlation; 1. A facial recognition system configured to store instructions to cause a facial recognition system to:

15. 15. The facial recognition system of claim 14, wherein the instructions are configured to cause the processing circuit to determine the correlation by comparing geometric facial features of the second facial image with geometric facial features of the facial identification data in the theme park database.

16. 15. The facial recognition system of claim 14, wherein the data in the plurality of amusement park area databases is copied from the theme park database.

17. The facial recognition system of claim 1 , wherein the second image capture device is the same as the third image capture device.

18. 10. The facial recognition system of claim 1, comprising a scanning station associated with the entrance to the area within the venue, the exit to the area within the venue, or both, the scanning station including the second image capture device, the third image capture device, or both.

19. a wearable device, wherein the computer system is configured to receive identification information of the wearable device from the scanning station, and the instructions direct the processing circuit to: receiving profile data associated with the identification information; linking the profile data to corresponding facial identification data based on the first facial image of the guest wearing the wearable device; 20. The facial recognition system of claim 18, wherein the facial recognition system is operable to:

20. 20. The facial recognition system of claim 19, wherein the profile data includes identification information of the guest, access information indicating access rights to features of the venue, or both.

Images