TWM603993U - Electronic authentication system - Google Patents

Abstract

An electronic authentication system is provided. The electronic authentication system includes a first touch apparatus, a storage apparatus, and a processor. The processor is configured to create a signature recognition model corresponding to a designated account through the first touch apparatus in advance to use the signature recognition model for authentication. Creation for the signature recognition model includes: receiving touch trajectory data through the first touch apparatus; identifying the touch trajectory data to obtain a plurality of trajectory features so as to create the signature recognition model based on the trajectory features; associating the signature recognition model with a designated account; and storing the signature recognition model to a location corresponding to the designated account in the storage apparatus.

Description

Electronic authentication system

The new creation is about an authentication system, and especially an electronic authentication system.

Traditionally, for transactions that require identification, signatures are usually used as identification. For example, in the case of bank transactions, when a customer handles business at the counter, he needs to sign multiple paper application forms, and then the bank clerk will manually check whether the signature matches the pre-saved signature before the transaction can be performed.

The new creation provides an electronic authentication system, which establishes a signature recognition model to automatically perform authentication.

The electronic authentication system created by the present invention includes: a first touch device, a storage device and a processor. The processor is coupled to the storage device and the first touch device. The processor is configured to pre-establish a signature recognition model corresponding to the designated account through the first touch device, so as to use the signature recognition model for authentication. Wherein, establishing a signature recognition model includes: receiving touch track data through the first touch device; The touch track data is distinguished to obtain a plurality of track characteristics, and the signature recognition model is established using the track characteristics; the signature recognition model is associated with a designated account; and the signature recognition model is stored in a storage device at a location corresponding to the designated account.

In an embodiment of the present invention, after the signature recognition model is established, the processor is configured to: receive the trace data to be verified through the first touch device; identify the trace data to be verified to obtain the characteristic data to be verified; Take the signature recognition model associated with the specified account; compare the characteristic data to be verified with the signature recognition model; if the characteristic data to be verified conforms to the signature recognition model, the verification is determined to be successful; and if the characteristic data to be verified does not conform to the signature recognition model, the verification is determined to be failed .

In an embodiment of the present invention, the aforementioned processor is configured to: receive the scanned document; perform text recognition on the scanned document to obtain an electronic text file; and store the electronic text file.

In an embodiment of the present invention, the above electronic authentication system further includes a scanning device. The scanning device scans the paper document to obtain the scanned document, and transmits the scanned document to the processor.

In an embodiment of the present invention, the aforementioned trajectory features include stroke order, writing trajectory, pen pressure, and time difference.

In an embodiment of the present invention, the above electronic authentication system further includes: a mobile device. The mobile device includes a second touch device to receive the trace data to be verified through the second touch device and transmit the trace data to be verified to the processor. The processor is configured to: receive the trace data to be verified; identify the trace data to be verified to obtain the characteristic data to be verified; read the signature recognition module associated with the designated account Type; compare the characteristic data to be verified with the signature recognition model; if the characteristic data to be verified conforms to the signature recognition model, the verification is determined to be successful; and if the characteristic data to be verified does not conform to the signature recognition model, the verification is determined to fail.

In an embodiment of the present invention, the above-mentioned processor is configured to: in the process of establishing the signature recognition model, receive the image including the seal pattern; recognize the seal pattern in the image to obtain the corresponding seal characteristic data; Associate the signature data with the designated account; and store the signature data in the storage device at a location corresponding to the designated account.

In an embodiment of the present invention, the above electronic authentication system further includes a touch tool. The touch tool is used for inputting touch track data in the first touch device.

Based on the above, a signature recognition model is established, and the signature recognition model is used to automatically recognize whether the customer’s handwritten signature matches the signature recognition model, which can save manpower and increase the work speed.

100, 200, 300: electronic authentication system

110: processor

120: The first touch device

130: storage device

201: Handwriting recognition module

202: Comparison module

203: Verification Database

204: Character recognition module

205: Document Database

210: electronic device

220: Touch Tools

310: mobile device

320: second touch device

Fig. 1 is a block diagram of an electronic authentication system according to an embodiment of the invention.

Fig. 2 is a block diagram of an electronic authentication system according to an embodiment of the invention.

Figure 3 is a block diagram of an electronic authentication system according to an embodiment of the invention Figure.

Fig. 1 is a block diagram of an electronic authentication system according to an embodiment of the invention. Please refer to FIG. 1, the electronic authentication system 100 includes a processor 110, a first touch device 120 and a storage device 130. The processor 110 is coupled to the first touch device 120 and the storage device 130. In an embodiment, the processor 110, the first touch device 120, and the storage device 130 may be provided in the same device at the same time. In another embodiment, the processor 110 and the storage device 130 are disposed in the same electronic device, and the first touch device 120 is a device different from the electronic device. In other embodiments, the processor 110, the first touch device 120, and the storage device 130 may also be independent devices.

The processor 110 is configured to pre-establish a signature recognition model corresponding to the designated account through the first touch device 120 to use the signature recognition model for authentication. Wherein, establishing a signature recognition model includes: receiving touch track data through the first touch device 120; recognizing the touch track data to obtain a plurality of track characteristics, using the track characteristics to establish a signature recognition model; and assigning the signature recognition model to The account is associated; and the signature recognition model is stored in the storage device 130 at a location corresponding to the designated account.

Fig. 2 is a block diagram of an electronic authentication system according to an embodiment of the invention. In FIG. 2, the electronic authentication system 200 includes an electronic device 210, a first touch device 120 and a touch tool 210. In this embodiment, the electronic device 210 and the first The touch device 120 is an independent component. In this embodiment, the user can perform touch operations on the first touch device 120 through the touch tool 210. In other embodiments, the user can also perform touch operations on the first touch device 120 through fingers. The processor 110 and the storage device 130 are provided in the electronic device 200.

The storage device 130 includes a handwriting recognition module 201, a comparison module 202 and a verification database 203. The processor 110 implements an identity authentication mechanism through the handwriting recognition module 201, the comparison module 202, and the verification database 203.

For example, for applications that require signature verification, a signature recognition model corresponding to a designated account can be created in advance for different users, so that the signature recognition model can be used for verification. First, the first touch device 120 receives touch track data and transmits the touch track data to the processor 110. After the processor 110 receives the touch trajectory data from the first touch device 120, it recognizes the touch trajectory data through the handwriting recognition module 201, thereby obtaining multiple trajectory features such as stroke sequence, writing trajectory, pen pressure, and time difference. . Then, a deep learning model is used to build a signature recognition model based on these trajectory features, and the signature recognition model is associated with the designated account, and then the signature recognition model is stored in the verification database 203 at a location corresponding to the designated account. The deep learning model is, for example, a convolutional neural network (CNN).

For example, each counter of a bank is provided with a first touch device 120 for the user to create files and perform verification. The first touch device 120 and the electronic device 210 can perform data transmission via wired transmission or wireless transmission.

After the file creation is completed, before the user wants to conduct various transactions next time, The handwritten signature is performed through the first touch device 120 again to obtain the trace data to be verified. After receiving the track data to be verified, the processor 110 recognizes the track data to be verified through the handwriting recognition module 201 to obtain the characteristic data to be verified, and then reads the signature recognition model associated with the designated account from the verification database 203. Then, the comparison module 202 compares the feature data to be verified with the signature recognition model. If the characteristic data to be verified conforms to the signature recognition model, it is determined that the verification is successful. If the characteristic data to be verified does not conform to the signature recognition model, it is determined that the verification has failed.

Fig. 3 is a block diagram of an electronic authentication system according to an embodiment of the invention. In FIG. 3, the electronic authentication system 300 includes an electronic device 210, a first touch device 120, a touch tool 210 and a mobile device 310. In this embodiment, the electronic device 210 further includes a character recognition module 204 and a document database 205.

The mobile device 310 includes a second touch device 320 to receive the trace data to be verified through the second touch device 320 and transmit the trace data to be verified to the processor 110. After the processor 110 receives the trace data to be verified, it recognizes the trace data to be verified through the handwriting recognition module 201 to obtain the feature data to be verified. Then, the comparison module 202 reads the signature recognition model associated with the specified account from the verification database 203, thereby comparing the signature recognition model to the characteristic data to be verified. If the characteristic data to be verified conforms to the signature recognition model, it is determined that the verification is successful. If the characteristic data to be verified does not conform to the signature recognition model, it is determined that the verification has failed.

In addition, the electronic authentication system 300 may further include a scanning device. The scanned document is obtained by scanning the paper document through the scanning device, and the scanned document is sent to the processor 110. After the processor 110 receives the scanned document, it scans the document through the character recognition module 204 Perform text recognition on the tracing document to obtain an electronic text file, and store the electronic text file in the document database 205.

In addition, in the process of establishing the signature recognition model, the processor 110 may also receive the image including the seal pattern, and further recognize the seal pattern in the image to obtain the corresponding seal feature data, and then combine the seal feature data with the designated account Associate, and then store the seal characteristic data to the location corresponding to the designated account in the verification database 203.

The above electronic authentication systems 100, 200, 300 can provide customers with the function of pre-filling the counter form online. Take cash transfer at the counter as an example. When conducting a transaction at the counter, it is necessary to print a flyer and ask the customer to sign and verify it. Through the above electronic authentication systems 100, 200, 300, a mobile device 310 (e.g. The tablet computer is for the customer to fill in the information, and after the customer confirms that it is correct, it directly signs on the second touch device 320 of the mobile device 310. After that, the system will automatically compare the client's signature with the signature recognition model of the verification database 203, and perform the transaction after confirming that it is correct. The financial side does not need to overprint the paper subpoena to ask the customer to sign, and there is no need to manually check the signature. The subsequent electronic documents can be directly archived after being reviewed by the supervisor.

In addition, when the clerk directly visits the customer to sign or guarantee, the current contract signing requires the customer to keep a contract with the bank. The clerk needs to scan and archive the documents afterwards, and use the mobile device 310 (such as The tablet computer) displays the contract and asks the customer to sign on the second touch device 320 of the mobile device 310. After that, the data received by the second touch device 320 can be automatically recorded and archived in the system, and there is no need to scan documents for archive.

To sum up, the new creation uses deep learning to establish a signature recognition model, and considers the sequence of strokes, writing trajectory, pen pressure and time difference when the client signs, and automatically recognizes whether the client’s signature matches the pre-established signature recognition model. , Which can save manpower and increase the operating speed.

100: Electronic authentication system

110: processor

120: The first touch device

130: storage device

Claims (8)

An electronic authentication system includes: a first touch device; a storage device; and a processor, coupled to the storage device and the first touch device, and configured to: through the first touch device Pre-establishing a signature recognition model corresponding to a designated account to use the signature recognition model for authentication, wherein establishing the signature recognition model includes: receiving a touch trace data through the first touch device; and recognizing the touch Trajectory data to obtain multiple trajectory features, use the trajectory features to build the signature recognition model; associate the signature recognition model with the designated account; and store the signature recognition model in the storage device corresponding to the designated account position. The electronic authentication system according to claim 1, wherein after the signature recognition model is established, the processor is configured to: receive a trace data to be verified through the first touch device; identify the trace data to be verified to obtain a The characteristic data to be verified; read the signature recognition model associated with the designated account; compare the characteristic data to be verified with the signature recognition model; if the characteristic data to be verified matches the signature recognition model, determine that the authentication is successful And if the feature data to be verified does not conform to the signature recognition model, it is determined that the verification fails. The electronic authentication system of claim 1, wherein the processor is configured to: receive a scanned document; perform text recognition on the scanned document to obtain an electronic text file; and store the electronic text file. The electronic authentication system according to claim 3 further includes: a scanning device, which scans a paper document to obtain the scanned document, and transmits the scanned document to the processor. The electronic authentication system according to claim 1, wherein the track characteristics include a stroke order, a writing track, a stroke pressure, and a time difference. The electronic authentication system according to claim 1, further comprising: a mobile device, including a second touch device, to receive a trace data to be verified through the second touch device, and transmit the trace data to be verified to the A processor configured to: receive the trace data to be verified; identify the trace data to be verified to obtain a feature data to be verified; read the signature recognition model associated with the designated account; compare the trace data to be verified Characteristic data and the signature recognition model; if the characteristic data to be verified matches the signature recognition model, it is determined that the verification is And if the feature data to be verified does not conform to the signature recognition model, it is determined that the verification fails. The electronic authentication system according to claim 1, wherein the processor is configured to: in the process of establishing the signature recognition model, receive an image including a seal pattern; identify the seal pattern in the image to obtain A corresponding seal feature data; associate the seal feature data with the designated account; and store the seal feature data in the storage device at a location corresponding to the designated account. The electronic authentication system according to claim 1, further comprising: a touch tool for inputting the touch track data in the first touch device.

Images