Image forming apparatus, method for controlling image forming apparatus, and storage medium

Abstract

To provide a mechanism capable of generating a file having an appropriate file name.SOLUTION: The MFP101 has a scan function of scanning a document. The MFP101 includes an image obtaining unit (CPU111) that obtains a scanned image of a document by executing the scan function, an inputting unit (the communication unit 126) that inputs the scanned image to the processing unit 307 capable of performing processing of generating a filename from the scanned image using a learning model, a filename obtaining unit (CPU111) that obtains the filename generated by the processing unit from the processing unit, and a generating unit (CPU111) that generates a file in which the scanned image is named with the filename obtained from the filename obtaining unit.SELECTED DRAWING: Figure 6

Description

[Technical Field]

[0001] The present invention relates to an image forming apparatus, a control method for an image forming apparatus, and a program. [Background technology]

[0002] In recent years, generation AI has become known as an artificial intelligence system. Generation AI can generate content such as text and images in response to prompts. Furthermore, generation AI's functionality can be expanded using plug-ins. For example, when a sentence containing the keyword "file name generation" is entered as an instruction to the generation AI, a plug-in related to file name generation is selected. The plug-in then performs processes that the generation AI has difficulty responding to, thereby enabling functional expansion. Furthermore, an image forming device with a scanning function can generate a file name from a scanned image and set the generated file name as the file name when saving the scanned image. Patent Document 1 discloses a device that performs character recognition processing (OCR processing) on ​​characters contained in a scanned image, displays a preview of the scanned image as the processing result, and generates a file name when the user selects a character string in the preview display. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 7150967 Summary of the Invention [Problem to be solved by the invention]

[0004] However, with the device described in Patent Document 1, when there is a character string to be selected when generating a file name, it can be difficult to select the character string depending on the position of the character string. For example, when the printed material to which a file name is to be set spans multiple pages, it can be difficult to select the character string if the character string to be used as the file name is on the first page. In this case, it may be difficult to generate a file with an appropriate file name.

[0005] The present invention has been made in view of the above-mentioned problems, and has an object to provide a mechanism that can generate files with appropriate file names. [Means for solving the problem]

[0006] In order to achieve the above object, the image forming apparatus of the present invention is an image forming apparatus having a scanning function for scanning a document, and is characterized by comprising: an image acquisition means for acquiring a scanned image of the document by executing the scanning function; an input means for inputting the scanned image to a processing unit capable of generating a file name from the scanned image using a learning model; a file name acquisition means for acquiring the file name generated by the processing unit from the processing unit; and a generation means for generating a file in which the scanned image is named with the file name acquired from the file name acquisition means. [Effects of the Invention]

[0007] According to the present invention, it is possible to generate files with appropriate file names. [Brief explanation of the drawings]

[0008] [Figure 1] 1 is a schematic diagram showing a configuration of an image forming system according to a first embodiment. [Figure 2] FIG. 2 is a block diagram showing an example of the hardware configuration of an MFP. [Figure 3] FIG. 2 is a block diagram showing an example of a hardware configuration of an information terminal. [Figure 4]FIG. 2 is a block diagram showing an example of the hardware configuration of a generation AI server. [Figure 5] FIG. 2 is a block diagram showing an example of a hardware configuration of a vector search server. [Figure 6] 10 is a flowchart showing a process executed by the information terminal. [Figure 7] 10 is a flowchart showing processing executed by the MFP. [Figure 8] This is a sequence diagram showing the processing executed between the information terminal, the MFP, the generation AI server, and the vector search server. [Figure 9] FIG. 10 is a diagram illustrating an example of a screen displayed on an operation unit of the information terminal. [Figure 10] FIG. 10 is a diagram showing an example of information about a document to be processed by the MFP. [Figure 11] A figure showing an example of information about a manuscript processed by a generation AI server. [Figure 12] FIG. 10 is a diagram showing an example of information about a manuscript processed by a vector search server. [Figure 13A] FIG. 10 is a diagram showing an example of information about a manuscript processed by a vector search server. [Figure 13B] FIG. 10 is a diagram showing an example of information about a manuscript processed by a vector search server. [Figure 14] FIG. 11 is a diagram showing an example of a screen displayed on an operation unit of an information terminal according to the second embodiment. [Figure 15] FIG. 10 is a diagram showing an example of information about a document to be processed by the MFP. [Figure 16] A figure showing an example of information about a manuscript processed by a generation AI server. [Figure 17] FIG. 10 is a diagram showing an example of information about a manuscript processed by a vector search server. [Figure 18A] FIG. 10 is a diagram showing an example of information about a manuscript processed by a vector search server. [Figure 18B] FIG. 10 is a diagram showing an example of information about a manuscript processed by a vector search server. DETAILED DESCRIPTION OF THE INVENTION

[0009] Each embodiment of the present invention will be described in detail below with reference to the drawings. However, the configurations described in each of the following embodiments are merely examples, and the scope of the present invention is not limited to the configurations described in each embodiment. For example, each component constituting the present invention can be replaced with any configuration that can perform the same function. Also, any component may be added. Furthermore, any two or more configurations (features) of each embodiment can be combined.

[0010] First Embodiment The first embodiment will be described below with reference to FIGS. 1 to 13B. FIG. 1 is a schematic diagram showing the configuration of an image forming system according to the first embodiment. As shown in FIG. 1, the image forming system 1000 includes an MFP 101, an information terminal 200, a generation AI server 300, and a vector search server 400, all of which are communicatively connected via a network 100. The MFP 101 is an image forming device having a printing function for printing on printing paper, a reading function for reading documents (i.e., a scanning function for scanning documents), and a fax function for sending and receiving still images such as characters and figures. Note that the image forming device is sufficient as long as it has at least a scanning function. The information terminal 200 is not particularly limited, and may be, for example, an information processing device such as a desktop or notebook personal computer, a tablet terminal, or a smartphone. The generation AI server 300 and the vector search server 400 are each, for example, an information processing device such as a desktop personal computer. Although the generation AI server 300 and the vector search server 400 are configured as separate entities in the configuration shown in Fig. 1, this is not limiting and, for example, they may be configured as an integrated entity. The network 100 may be, for example, the Internet or a LAN (Local Area Network). Furthermore, the network 100 may be either a wired or wireless network.

[0011] Fig. 2 is a block diagram showing an example of the hardware configuration of an MFP. As shown in Fig. 2, MFP 101 has control unit 110, operation unit 116, reading unit (image acquisition means) 118, printing unit 120, wireless communication unit 122, communication unit 126, FAX communication unit 124, and communication unit 126, which are connected to each other so as to be able to communicate with each other. Control unit 110 has CPU 111, ROM 112, RAM 113, storage 114, operation unit I / F 115, reading unit I / F 117, printing unit I / F 119, wireless communication unit I / F 121, communication unit I / F 125, and USB_I / F 127. Note that the configuration of control unit 110 is not limited to the configuration shown in Fig. 2, and may have, for example, multiple CPUs 111, ROMs 112, and RAMs 113.

[0012] The control unit 110 is a computer that controls the overall operation of the MFP 101. The CPU 111 loads a control program stored in the ROM 112 or storage 114 into the RAM 113 to perform, for example, scan job analysis, reading control, and printing control. The ROM 112 stores, for example, a control program executable by the CPU 111, as well as a boot program and font data. The control program includes, for example, a program that causes the control unit 110 to execute each unit and each means (control method of the image forming apparatus) of the image forming apparatus. The RAM 113 is the main memory of the CPU 111. The RAM 113 is used, for example, as a work area, an auxiliary information storage unit that stores auxiliary information (supplementary information) described below, and a temporary storage area for expanding the control program. The storage 114 stores, for example, auxiliary information from the information terminal 200, reading instruction information, request information to the generation AI server 300, file name suggestions from the generation AI server 300, print data, various programs, and various setting information. In this embodiment, a flash memory is used as the storage 114, but this is not limiting. For example, the storage 114 may be a Solid State Drive (SSD), a Hard Disk Drive (HDD), an embedded Multi Media Card (eMMC), or the like.

[0013] An operation unit 116 is connected to the operation unit I / F 115. The operation unit 116 displays information to a user using the MFP 101 and receives instructions from the user. The operation unit 116 is not particularly limited and may be, for example, a keyboard, a mouse, or a touch panel. A reading unit 118 is connected to the reading unit I / F 117. The reading unit 118 reads an original document and converts the image of the original document into image data such as binary data. This image data is transmitted to an external device such as the generation AI server 300 or the information terminal 200, or printed on printing paper. The reading unit 118 also has a feeder (ADF) that transports the original document to be read. A printing unit 120 is connected to the printing unit I / F 119. The CPU 111 transfers image data (print data) stored in the RAM 113 to the printing unit 120 via the printing unit I / F 119. The printing unit 120 prints this image data on printing paper fed from a paper feed cassette. A wireless communication unit 122 is connected to the wireless communication unit I / F 121. This allows wireless communication with external wireless devices via the wireless communication unit 122. A fax communication unit 124 is connected to the fax unit I / F 123. The fax communication unit 124 is also connected to the public line network 107. This allows still images such as characters and figures to be sent and received via the fax communication unit 124. A communication unit 126 is connected to the communication unit I / F 125. The communication unit 126 is also connected to the network 100. This allows various information such as image data to be sent to and received from external devices on the network 100 via the communication unit 126. A USB memory 128 is connected to the USB_I / F 127. For example, a file name is assigned to the image data generated by the reading unit 118 and the image data is stored in the USB memory 128. The image forming system 1000 may transmit and receive data via the network 100 using, for example, email, FTP, SMB, WEBDAV, or other protocols.Furthermore, the MFP 101 may transmit and receive scan jobs and various setting data via the network 100 by accessing the generation AI server 300 via HTTP communication.

[0014] FIG. 3 is a block diagram showing an example of the hardware configuration of an information terminal. In FIG. 3, the information terminal 200 includes an operation panel 201, a camera 204, an NFC communication unit 205, a Bluetooth (registered trademark) communication unit 206, a CPU 207, a ROM 208, a RAM 209, a storage 210, and a wireless LAN communication unit 211. The CPU 207 loads a control program stored in the ROM 208 into the RAM 209 and executes various processes. The ROM 208 stores, for example, a control program executable by the CPU 207. The RAM 209 is the main memory of the CPU 207. The RAM 209 is used as a work area and as a temporary storage area for loading the control program. The storage 210 stores various data, such as photographs and electronic documents. The operation panel 201 has a touch panel function capable of detecting touch operations by a user of the information terminal 200. This allows desired operation instructions to be input to the information terminal 200. The operation panel 201 also displays various screens provided by the OS and an email sending application. The information terminal 200 may have a keyboard (not shown). This keyboard can be used to input desired operation instructions to the information terminal 200. The camera 204 captures an image of a subject. This image is stored, for example, in a predetermined area of ​​the storage 210. The camera 204 can capture an image of a QR code (registered trademark). Then, information contained in the QR code can be obtained using a program capable of analyzing the QR code. The information terminal 200 can transmit and receive data to and from various peripheral devices via the NFC communication unit 205, the Bluetooth communication unit 206, and the wireless LAN communication unit 211. It is preferable that the Bluetooth communication unit 206 is compatible with Bluetooth Low Energy. A setup tool for the MFP 101 is installed (saved) in a predetermined area of ​​the storage 210. This setup tool is deployed in the RAM 209 and executed.

[0015] FIG. 4 is a block diagram showing an example of the hardware configuration of the generation AI server. As shown in FIG. 4, the generation AI server 300 includes a CPU 301, a ROM 302, a RAM 303, a storage 304, a communication unit 305, and a GPU 306. The CPU 301 loads a control program stored in the ROM 302 or the storage 304 into the RAM 303 and executes various processes, such as image recognition, character recognition, and a registered file name suggestion plug-in. The ROM 302 stores, for example, a control program executable by the CPU 301 and other programs, such as a boot program. The RAM 303 is the main memory of the CPU 301. The RAM 303 is used, for example, as a work area or a temporary storage area for loading the control program. The storage 304 stores, for example, image data, auxiliary information, and request information from the MFP 101, search information for the vector search server 400, various programs, and various setting information. The storage 304 also stores search result information from the vector search server 400, response information for the MFP 101, and the like. In this embodiment, flash memory is used as the storage 304, but this is not limiting. For example, an SSD, HDD, eMMC, or the like may be used as the storage 304. The communication unit 305 is connected to the network 100. This allows communication with, for example, the vector search server 400. The GPU 306 is used, for example, for neural network calculations. Note that the generation AI server 300 may be equipped with a TPU or the like instead of the GPU 306. Furthermore, the generation AI server 300 may be, for example, a virtual server operating within a single piece of hardware.

[0016] FIG. 5 is a block diagram showing an example of the hardware configuration of a vector search server. As shown in FIG. 5, the vector search server 400 includes a CPU 401, a ROM 403, a RAM 402, a storage 404, a communication unit 405, and a GPU 406. The CPU 401 loads a control program stored in the ROM 403 or storage 404 into the RAM 402. The CPU 401 then executes various processes, such as vectorization, vector synthesis, vector search, and character conversion, to perform vector search processing according to search information from the generation AI server 300. The ROM 403 stores, for example, a control program executable by the CPU 401 and other programs, such as a boot program. The RAM 402 is the main memory of the CPU 401. The RAM 402 is used, for example, as a work area or a temporary storage area for expanding the control program. The storage 404 stores search information from the generation AI server 300, search result information to the generation AI server 300, vector DB information for vector search, various programs, and various setting information. In this embodiment, flash memory is used as the storage 404, but this is not limiting. For example, an SSD, HDD, eMMC, or the like may be used as the storage 404. The communication unit 405 is connected to the network 100. This allows communication with, for example, the generation AI server 300. The GPU 406 is used, for example, for neural network calculations. Note that the vector search server 400 may be equipped with a TPU or the like instead of the GPU 406. Furthermore, the vector search server 400 may be, for example, a virtual server operating within a single piece of hardware.

[0017] Fig. 6 is a flowchart showing the processing executed by the information terminal. As shown in Fig. 6, in step S701, the CPU 207 of the information terminal 200 displays a screen 901 (see Fig. 9(a)) on the operation panel 201, on which auxiliary information 902 can be input, and accepts input of the auxiliary information 902 on the screen 901. The "auxiliary information" refers to information that assists in the generation of a file name for a scanned image when the file name is generated by the processing unit 307, which will be described later.

[0018] In step S702, the CPU 207 accepts an operation on a read start button 906 (see FIG. 9B) included in the screen 901. The read start button 906 is a button for instructing the MFP 101 to start reading the document.

[0019] In step S703, the CPU 207 transmits an instruction to start reading the document to the MFP 101. This instruction also includes an instruction to execute a proposal process that proposes a file name.

[0020] In step S704, CPU 207 determines whether a file name proposal has been received from MFP 101. If it is determined in step S704 that a file name proposal has been received, the process proceeds to step S705. On the other hand, if it is determined in step S704 that a file name proposal has not been received, the process remains in step S704 and waits.

[0021] In step S705, the CPU 207 displays the file name (see FIG. 9(c)) on the operation panel 201.

[0022] In step S706, if the save button 908 (see FIG. 9C) included in the screen 901 is operated, the CPU 207 approves the file name displayed in step S705, and the process ends.

[0023] FIG. 7 is a flowchart showing processing executed by the MFP. As shown in FIG. 7, in step S801, the CPU 111 of the MFP 101 acquires auxiliary information 902 from the information terminal 200. In this manner, in this embodiment, the CPU 111 also functions as an information acquisition unit that acquires the auxiliary information. Note that in the MFP 101, a section that functions as an information acquisition unit may be provided separately from the CPU 111. The CPU 111 stores the auxiliary information 902 as auxiliary information 1002 (see FIG. 10(b)) in the storage 114. This auxiliary information 1002 includes first auxiliary information 1003 and second auxiliary information 1004. The first auxiliary information 1003 is the character string "Japanese". The second auxiliary information 1004 is the character string "proposal".

[0024] In step S802, CPU 111 determines whether or not an instruction to start reading a document has been received from information terminal 200. If it is determined in step S802 that an instruction to start reading a document has been received, the process proceeds to step S803. On the other hand, if it is determined in step S802 that an instruction to start reading a document has not been received, the process remains in step S802 and waits.

[0025] In step S803, CPU 111 controls reading unit 118, i.e., executes the scan function, to read the document and acquire a scanned image of the document, i.e., scanned image 1001 (see FIG. 10(a)) (image acquisition step). In this manner, in this embodiment, CPU 111 also functions as an image acquisition means for acquiring scanned image 1001. Note that in MFP 101, a section that functions as an information acquisition means may be provided separately from CPU 111. Then, scanned image 1001 is stored in storage 114. Scanned image 1001 includes an image of ramen, an image of the character string "BUSINESS," and an image of the character string "PROPOSAL."

[0026] In step S804, the CPU 111 transmits the scanned image 1001, auxiliary information 1002, and input script 1005 (see FIG. 10(c)) stored in the storage 114 as a file name proposal request to the generation AI server 300 from the communication unit 126.

[0027] In step S805, the CPU 111 determines whether a file name proposal has been received from the generation AI server 300. If it is determined in step S805 that a file name proposal has been received (file name acquisition step), the process proceeds to step S806. The CPU 111 then stores the file name proposed by the generation AI server 300 (see FIG. 10(d)) in the storage 114. On the other hand, if it is determined in step S805 that a file name proposal has not been received, the process remains in standby at step S805. In this manner, in this embodiment, the CPU 111 also functions as a file name acquisition means that acquires a file name. Note that in the MFP 101, a section that functions as a file name acquisition means may be provided separately from the CPU 111.

[0028] In step S806, CPU 111 instructs information terminal 200 to display the file name stored in step S805.

[0029] In step S807, CPU 111 accepts the file name approval from information terminal 200.

[0030] In step S808, CPU 111 generates a file by assigning the file name approved in step S807 to scanned image 1001 acquired in step S803 (generation step). In this manner, in this embodiment, CPU 111 also functions as a generation unit that generates a file. Note that in MFP 101, a section that functions as a generation unit may be provided separately from CPU 111. This file is then stored (saved) in USB memory 128 connected to USB_I / F 127. Note that the file may be stored in storage 114 depending on the setting state of the scan job, or may be transmitted to an external device on network 100 via communication unit I / F 125 and communication unit 126.

[0031] 8 is a sequence diagram showing the processing executed between the information terminal, the MFP, the generation AI server, and the vector search server. As shown in FIG. 8, in step S601, the information terminal 200 transmits (inputs) auxiliary information 1002 to the MFP 101.

[0032] In step S602, information terminal 200 (CPU 207) instructs MFP 101 to execute a reading process for reading a document. This instruction includes execution of a file name suggestion process for suggesting a file name.

[0033] In step S603, MFP 101 (CPU 111) executes a reading process using reading unit 118. As a result, a scanned image (image data) of the original is acquired and stored in storage 114. Note that, although it depends on the original, this scanned image contains at least one of information such as characters, symbols, figures, and photographs. Also, in step S603, if the original is a printed matter having multiple pages, a scanned image of the first page of the printed matter is acquired as the scanned image. This is because, among the multiple pages, the first page tends to contain a relatively large amount of information for generating a file name. In this embodiment, the scanned image acquired in step S603 is scanned image 1001 shown in FIG. 10(a).

[0034] In step S604, the MFP 101 instructs (requests) the generation AI server 300 to execute a file name proposal process. This instruction includes a file name proposal request 1101 shown in FIG. 11(a). The file name proposal request 1101 includes, for example, a scanned image 1102 and an input prompt 1103. The scanned image 1102 is the scanned image 1001 acquired in step S603. The input prompt 1103 includes auxiliary information 1002 and an input script 1005, which is instruction information instructing the generation of a file name. In step S604, the file name proposal request 1101 is input to the generation AI server 300 (input step). This input is performed by the communication unit 126 of the MFP 101. As described above, in this embodiment, the communication unit 126 functions as an input unit. The file name proposal request 1101 is input to a processing unit 307 built into the generation AI server 300. Processing unit 307 is configured with CPU 301 and GPU 306, and is capable of executing a process of generating a file name from file name proposal request 1101 using a learning model. Processing unit 307 is also capable of executing image recognition processing and character recognition processing. Note that, although processing unit 307 is capable of executing image recognition processing and character recognition processing in this embodiment, it is not limited thereto, and it is sufficient if it is capable of executing at least one of image recognition processing and character recognition processing.

[0035] In step S605, the generation AI server 300 (processing unit 307) extracts the input script 1005 from the input prompt 1103 of the file name proposal request 1101. This input script 1005 is stored, for example, in the storage 304 as the input script 1111 (see FIG. 11(d)). The generation AI server 300 interprets the input script 1111 stored in the storage 304. As a result of this interpretation, it is determined that a file name proposal is being requested, and the use of the file name proposal plug-in is selected. The generation AI server 300 performs image recognition processing and character recognition processing on the scanned image 1102 included in the file name proposal request 1101. As a result, a group of characters 1104 is obtained from the scanned image 1102. The group of characters 1104 includes the character strings 1105, 1106, and 1107. The character string 1105 is the character string "ramen" obtained by image recognition processing on the scanned image 1102. In this embodiment, the processing unit 307 has a trained model that has been trained in advance for an image of ramen, and when a new image of ramen is input, the processing unit 307 can output the character string "Ramen" using the trained model. A character string 1106 is the character string "Business" obtained by character recognition processing on the scanned image 1102. A character string 1107 is the character string "Proposal" obtained by character recognition processing on the scanned image 1102. Furthermore, a character string group 1108 is acquired from the input prompt 1103 by character recognition processing on the scanned image 1102. The character string group 1108 is stored in the storage 304. The character string group 1108 includes a character string 1109 and a character string 1110. The character string 1109 is the character string "Japanese." The character string 1110 is the character string "Proposal." The generation AI server 300 generates search information 1112 (see FIG. 11(e)) based on the character string group 1104 and the character string group 1108. The search information 1112 includes first information 1113, second information 1114, and third information 1115. The first information 1113 is a word group "Ramen Japanese Proposal." The second information 1114 is a word group "Business Japanese Proposal." The third information 1115 is a word group "Proposal Japanese Proposal."

[0036] In step S606, the generation AI server 300 transmits search information 1112 to the vector search server 400 as a corresponding word search request for searching for corresponding words.

[0037] In step S607, the vector search server 400 performs a corresponding word search using a known technique such as Word2Vec. This corresponding word search is performed by the processing unit 407, which is composed of the CPU 401 and the GPU 406. In Word2Vec, the meaning of a word is expressed, for example, by a multidimensional vector 1222 (see FIG. 12(c)) composed of an array of multiple numerical values. Furthermore, the storage 404 of the vector search server 400 pre-stores search information 1201. The search information 1201 includes a corresponding word search request 1202, first information 1203, second information 1204, and third information 1205 (see FIG. 12(a)). The first information 1203 is a word group "Ramen Japanese Proposal" corresponding to the consensus of the search word. The second information 1204 is a word group "Business Japanese Proposal" corresponding to the consensus of the search word. The third information 1205 is a word group "Proposal Japanese Proposal" corresponding to the consensus of the search word.

[0038] The processing unit 407 performs multidimensional vectorization of the words included in each of the first information 1203 to the third information 1205. As a result of this multidimensional vectorization, for example, a vectorization 1206 (see FIG. 12(b)) is obtained. In the vectorization 1206, the character string 1207 (Ramen) is converted into the character string 1209 (V_Ramen). The character string 1210 (Japanese) is converted into the character string 1212 (V_Japanese). The character string 1213 (Proposal) is converted into the character string 1215 (V_Proposal). The character string 1216 (Business) is converted into the character string 1218 (V_Business). The character string 1219 (Proposal) is converted into the character string 1221 (V_Proposal).

[0039] Next, the processing unit 407 acquires a multidimensional vector for search. As a result of this acquisition, for example, a vector combination 1301 (see FIG. 13A(a)) is obtained. In the vector combination 1301, a multidimensional vector 1303 (V_1) is acquired from a composite character string 1302 (V_Ramen + V_Japanese + V_Proposal) obtained by combining the character strings 1209, 1212, and 1215. The multidimensional vector 1303 corresponds to the first information 1203 (a group of words "Ramen Japanese Proposal"). Similarly, a multidimensional vector 1305 (V_2) is acquired from a composite character string 1304 (V_Business + V_Japanese + V_Proposal) obtained by combining the character strings 1218, 1212, and 1215. The multidimensional vector 1303 corresponds to the second information 1204 (a group of words "Business Japanese Proposal"). Furthermore, a multidimensional vector 1307 (V_3) is obtained from a composite character string 1306 (V_Proposal + V_Japanese + V_Proposal) that is a composite of the character strings 1221, 1212, and 1215. The multidimensional vector 1307 corresponds to the third information 1205 (a group of words “Proposal Japanese Proposal”).

[0040] Next, the processing unit 407 performs a search within the vector DB information stored in the storage 404. As a result of this search, for example, a vector search 1308 (see FIG. 13A(b)) is obtained. In the vector search 1308, the similarity between the multidimensional vector 1303 (V_1) and the vector DB information is compared to search for the most similar word, and a character string 1209 (V_Ramen) is obtained. Similarly, the similarity between the multidimensional vector 1305 (V_2) and the vector DB information is compared to search for the most similar word, and a multidimensional vector 1311 (V_Business) is obtained. Furthermore, the similarity between the multidimensional vector 1307 (V_3) and the vector DB information is compared to search for the most similar word, and a multidimensional vector 1313 (V_Proposal) is obtained. Note that in this embodiment, the cosine similarity that obtains the similarity between multidimensional vectors is used for the similarity, but the present invention is not limited to this.

[0041] Next, the processing unit 407 converts the multidimensional vector into a general character string. As a result of this conversion, for example, character conversion 1314 (see FIG. 13A(c)) is obtained. In character conversion 1314, character string 1207 (ramen) is obtained from character string 1209 (V_ramen). Similarly, character string 1317 (business) is obtained from multidimensional vector 1311 (V_business). Furthermore, character string 1319 (proposal) is obtained from multidimensional vector 1312 (V_proposal).

[0042] In step S608, the vector search server 400 transmits the corresponding word search result in step S607 to the generation AI server 300. This corresponding word search result is, for example, corresponding word search result 1801 (see FIG. 13B(a)). Corresponding word search result 1801 includes character strings 1802, 1803, and 1804. Character string 1802 is the character string "ramen." Character string 1803 is the character string "business." Character string 1804 is the character string "proposal." Corresponding word search result 1801 is stored in storage 304.

[0043] In step S609, the generation AI server 300 (processing unit 307) generates a file name 1805 (see FIG. 13B(b)) from character strings 1802 to 1804 in the corresponding word search result 1801. The file name 1805 is the file name "Ramen Business Proposal." Note that depending on how character strings 1802 to 1804 are arranged, file name 1805 could be, for example, "Business Ramen Proposal" or "Business Proposal Ramen," but the learning model of the processing unit 307 has been trained to generate a natural file name such as "Ramen Business Proposal."

[0044] In step S610, the generation AI server 300 transmits a file name proposal 1006 (see FIG. 10(d)) to the MFP 101. The file name proposal 1006 includes a file name 1805 (ramen business proposal). The file name proposal 1006 is also acquired by the communication unit 126 of the MFP 101 and stored, for example, in the storage 114. In this manner, in this embodiment, the MFP 101 also functions as a file name acquisition means for acquiring the file name 1805.

[0045] In step S611, the MFP 101 transmits the file name proposal 1006 stored in the storage 114 to the information terminal 200, and instructs the information terminal 200 to display the file name proposal 1006 on the operation panel 201.

[0046] In step S612, information terminal 200 approves the file name "Ramen Business Proposal" included in file name proposal 1006 displayed on operation panel 201 as the file name of scanned image 1001. Then, this approval information is transmitted to MFP 101.

[0047] In step S613, prior to file generation, which will be described later, MFP 101 determines whether to name scanned image 1001 with the file name "Ramen Business Proposal" included in file name proposal 1006, based on the approval information in step S612. As described above, the approval information in step S612 is information indicating that the file name "Ramen Business Proposal" is approved as the file name of scanned image 1001. Therefore, in step S613, it is determined that scanned image 1001 should be named with the file name "Ramen Business Proposal." Then, once the naming is determined, MFP 101 generates a file (e.g., a PDF file) in which scanned image 1001 is named with the file name "Ramen Business Proposal." The file generated in this manner has an appropriate file name, i.e., an accurate file. Note that file generation is performed by CPU 111. As described above, in this embodiment, CPU 111 also functions as a generating unit that generates a file. In MFP 101, a unit that functions as a generating unit may be provided separately from CPU 111. In addition, the file may be stored (saved) in the storage 114, stored in a USB memory 128 connected to the USB_I / F 127, or transmitted to another external device via the communication unit 126, for example.

[0048] FIG. 9 is a diagram showing an example of a screen displayed on the operation unit of an information terminal. Screen 901 shown in FIG. 9(a) is an input screen on which auxiliary information 902 can be input. The auxiliary information 902 contributes to improving the accuracy of file name suggestions (file name generation) by the generation AI server 300. The auxiliary information 902 includes first auxiliary information 903 and second auxiliary information 904. The first auxiliary information 903 is the name of the language used in the file name of the scanned image 1001, which may be, for example, "Japanese" in this embodiment. The second auxiliary information 904 is information included in the scanned image 1001, which may be, for example, "proposal" in this embodiment. Screen 901 also includes a setting button 905. By operating the setting button 905 after inputting the auxiliary information 902, the auxiliary information 902 is transmitted to the MFP 101.

[0049] Screen 901 also includes a start reading button 906 shown in FIG. 9(b). Operating start reading button 906 can instruct the reading unit 118 of MFP 101 to start reading the document. Screen 901 also displays a file name suggestion result 907 and a save button 908 shown in FIG. 9(c) after a file name is suggested. File name suggestion result 907 is a message regarding the file name suggested by the generation AI server 300. Operating save button 908 saves a file with the file name assigned to the scanned image 1001.

[0050] FIG. 10 is a diagram showing an example of information related to a document processed by the MFP. A scanned image 1001 shown in FIG. 10(a) is a scanned image obtained by scanning a document using the scanning function of the reading unit 118 of the MFP 101. Auxiliary information 1002 shown in FIG. 10(b) is similar to the auxiliary information 902 and includes first auxiliary information 1003 and second auxiliary information 1004. Like the first auxiliary information 903, the first auxiliary information 1003 is the name of the language used in the file name of the scanned image 1001, "Japanese." Like the second auxiliary information 904, the second auxiliary information 1004 is the information "proposal" included in the scanned image 1001. An input script 1005 shown in FIG. 10(c) is instruction information that instructs the generation AI server 300 to generate a file name. The input script 1005 is pre-stored in the storage 114 of the MFP 101. The file name proposal 1006 shown in Figure 10(d) is sent from the generation AI server 300 and is stored in the storage 114 after being sent. By changing the instruction information in the input script 1005, it is possible to propose, in addition to the file name proposal, for example, a path name including a folder name, or a relatively long sentence as a file property.

[0051] FIG. 11 shows an example of information about a document processed by the generation AI server. The filename proposal request 1101 shown in FIG. 11(a) is an instruction from the MFP 101 to the generation AI server 300 to execute a filename proposal process. The filename proposal request 1101 includes a scanned image 1102 and an input prompt 1103. The scanned image 1102 is similar to the scanned image 1001. The input prompt 1103 includes the auxiliary information 1002 and input script 1005 described above. The string group 1104 shown in FIG. 11(b) is obtained by image recognition and character recognition processing on the scanned image 1102. This string group 1104 includes strings 1105, 1106, and 1107. The string 1105 is the string "Ramen" obtained by image recognition processing on the scanned image 1102. The string 1106 is the string "Business" obtained by character recognition processing on the scanned image 1102. A character string 1107 is the character string "Proposal" obtained by character recognition processing of the scanned image 1102. A character string group 1108 shown in FIG. 11(c) is obtained from the input prompt 1103. The character string group 1108 includes a character string 1109 (Japanese) and a character string 1110 (proposal), similar to the auxiliary information 1002. An input script 1111 shown in FIG. 11(d) is obtained from the input prompt 1103. The input script 1111 is similar to the input script 1005. Search information 1112 shown in FIG. 11(e) is generated based on the character string group 1104 and the character string group 1108. This search information 1112 includes first information 1113 (Ramen Japanese Proposal), second information 1114 (Business Japanese Proposal), and third information 1115 (Proposal Japanese Proposal).

[0052] FIG. 12 shows an example of information related to a manuscript processed by the vector search server. Search information 1201 shown in FIG. 12(a) is notified from the generation AI server 300. This search information 1201 includes a corresponding word search request 1202 indicating a similarity search, first information 1203 (ramen Japanese proposal), second information 1204 (business Japanese proposal), and third information 1205 (business Japanese proposal). Vectorization 1206 shown in FIG. 12(b) is the result of multidimensional vectorization executed by the processing unit 407. As described above, character string 1207 (ramen) is converted to character string 1209 (V_ramen). Character string 1210 (Japanese) is converted to character string 1212 (V_Japanese). Character string 1213 (proposal) is converted to character string 1215 (V_proposal). Character string 1216 (Business) is converted to character string 1218 (V_Business). The character string 1219 (Proposal) is converted into a character string 1221 (V_Proposal). In the multidimensional vector 1222 shown in Fig. 12(c), the meaning of a word obtained by the corresponding word search is composed of an array of multiple numerical values.

[0053] FIG. 13A is a diagram showing an example of information about a manuscript processed by a vector search server. A vector composition 1301 shown in FIG. 13A(a) is the result obtained by multidimensional vectorization for search. In the vector composition 1301, a composite character string 1302 (V_Ramen + V_Japanese + V_Proposal) is composed of character strings 1209, 1212, and 1215. A multidimensional vector 1303 (V_1) is obtained from the composite character string 1302. Similarly, a composite character string 1304 (V_Business + V_Japanese + V_Proposal) is composed of character strings 1218, 1212, and 1215. A multidimensional vector 1305 (V_2) is obtained from the composite character string 1304. Furthermore, a composite character string 1306 (V_Proposal + V_Japanese + V_Proposal) is composed of character strings 1221, 1212, and 1215. From the composite string 1306, a multidimensional vector 1307 (V_3) is obtained.

[0054] Vector search 1308 shown in FIG. 13A(b) is a search result in the vector DB information stored in storage 404. In vector search 1308, the similarity between multidimensional vector 1303 and vector DB information is compared, and character string 1209 (V_Ramen) is obtained. Similarly, the similarity between multidimensional vector 1305 and vector DB information is compared, and multidimensional vector 1311 (V_Business) is obtained. Furthermore, the similarity between multidimensional vector 1307 and vector DB information is compared, and multidimensional vector 1313 (V_Proposal) is obtained. Character conversion 1314 shown in FIG. 13A(c) is a conversion result when a multidimensional vector is converted into a character string. In character conversion 1314, character string 1207 (Ramen) is obtained from character string 1209. Similarly, character string 1317 (Business) is obtained from multidimensional vector 1311. Furthermore, character string 1319 (Proposal) is obtained from multidimensional vector 1313.

[0055] Fig. 13B is a diagram showing an example of information related to a manuscript processed by the vector search server. Corresponding word search result 1801 shown in Fig. 13B(a) is the corresponding word search result in step S607. Corresponding word search result 1801 includes character string 1802 (ramen), character string 1803 (business), and character string 1804 (proposal). File name 1805 shown in Fig. 13B(b) is generated from character strings 1802 to 1804.

[0056] Second Embodiment The second embodiment will be described below with reference to FIGS. 14 to 18B. Differences from the previous embodiment will be mainly described, and similar details will not be described again. This embodiment is similar to the first embodiment except for the type of document to be scanned. FIG. 14 illustrates an example of a screen displayed on the operation unit of an information terminal according to the second embodiment. A screen 1411 shown in FIG. 14(a) is an input screen on which auxiliary information 1412 can be input. The auxiliary information 1412 includes first auxiliary information 1413 and second auxiliary information 1414. The first auxiliary information 1413 is a word (information) indicating the level of detail in the file name of the scanned image 1401 (see FIG. 15(a)). For example, in this embodiment, it can be "concise." The second auxiliary information 1414 is a word (information) indicating the purpose of the file. For example, in this embodiment, it can be "proposal." The screen 1411 also includes a setting button 1415. By operating the set button 1415 with the auxiliary information 1412 entered, the auxiliary information 1412 is sent to the MFP 101. In addition, after the file name is proposed, the screen 1411 displays a file name proposal result 1417 and a save button 1418 shown in FIG. 14(b). The file name proposal result 1417 is a message regarding the file name proposed by the generation AI server 300. By operating the save button 1418, a file with the file name assigned to the scanned image 1401 is saved.

[0057] FIG. 15 illustrates an example of information related to a document processed by an MFP. The scanned image 1401 shown in FIG. 15(a) is a scanned image obtained by scanning a document using the scanning unit 118 of the MFP 101. The auxiliary information 1402 shown in FIG. 15(b) is similar to the auxiliary information 1412 and includes first auxiliary information 1403 and second auxiliary information 1404. Like the first auxiliary information 1413, the first auxiliary information 1403 is the word "concise," indicating the level of detail in the filename of the scanned image 1401. Like the second auxiliary information 1414, the second auxiliary information 1404 is the word "proposal," indicating the purpose of the file. The input script 1405 shown in FIG. 15(c) is instruction information instructing the generation AI server 300 to generate a filename. The filename proposal 1406 shown in FIG. 15(d) is stored in the storage 114 after being sent from the generation AI server 300.

[0058] FIG. 16 is a diagram showing an example of information related to a manuscript processed by the generation AI server. File name proposal request 1501 shown in FIG. 16(a) includes scanned image 1502 and input prompt 1503. Scanned image 1502 is similar to scanned image 1401. Input prompt 1503 includes auxiliary information 1402 and input script 1405. Character string group 1504 shown in FIG. 16(b) includes character strings 1505, 1506, and 1507. Character string 1505 is the character string "Single-lens reflex camera" obtained by image recognition processing on scanned image 1502. Character string 1506 is the character string "Classic design" obtained by character recognition processing on scanned image 1502. Character string 1507 is the character string "Digital camera" obtained by character recognition processing on scanned image 1502. A character string group 1508 shown in Fig. 16(c) includes a character string 1509 (concise) and a character string 1510 (proposal), similar to the auxiliary information 1402. An input script 1511 shown in Fig. 16(d) is acquired from an input prompt 1503. The input script 1511 is similar to the input script 1405. Search information 1512 shown in Fig. 16(e) includes first information 1513 (single-lens reflex camera concise proposal), second information 1514 (classic concise proposal), and third information 1515 (digital camera concise proposal).

[0059] FIG. 17 shows an example of information related to a manuscript processed by a vector search server. Search information 1601 shown in FIG. 17(a) includes a corresponding word search request 1602, first information 1603 (SLR camera, concise, proposal), second information 1604 (classic, concise, proposal), and third information 1605 (digital camera, concise, proposal). Vectorization 1606 shown in FIG. 17(b) is the result of multidimensional vectorization. In vectorization 1606, character string 1607 (SLR camera) is converted into character string 1609 (V_SLR camera). Character string 1610 (concise) is converted into character string 1612 (V_concise). Character string 1613 (proposal) is converted into character string 1615 (V_proposal). The character string 1616 (Classic Design) is converted to the character string 1618 (V_(Classic Design)). The character string 1619 (Digital Camera) is converted to the character string 1621 (V_Digital Camera).

[0060] FIG. 18A is a diagram showing an example of information related to a manuscript processed by a vector search server. In vector synthesis 1701 shown in FIG. 18A(a), a synthesized character string 1702 (V_SLR camera + V_succinct + V_proposal) is synthesized with character strings 1609, 1612, and 1615. A multidimensional vector 1703 (V_21) is obtained from the synthesized character string 1702. Similarly, a synthesized character string 1704 (V_classic design + V_succinct + V_proposal) is synthesized with character strings 1618, 1612, and 1615. A multidimensional vector 1705 (V_22) is obtained from the synthesized character string 1704. Furthermore, a synthesized character string 1706 (V_digital camera + V_succinct + V_proposal) is synthesized with character strings 1621, 1612, and 1615. From the composite string 1706, a multidimensional vector 1707 (V_23) is obtained.

[0061] In vector search 1708 shown in FIG. 18A(b), the similarity between multidimensional vector 1703 and vector DB information is compared, and multidimensional vector 1709 (V_SLR) is obtained. Similarly, the similarity between multidimensional vector 1705 and vector DB information is compared, and multidimensional vector 1711 (V_Classical) is obtained. Furthermore, the similarity between multidimensional vector 1707 and vector DB information is compared, and multidimensional vector 1713 (V_Digital Camera) is obtained. In character conversion 1714 shown in FIG. 18A(c), a character string 1720 (SLR) is obtained from multidimensional vector 1709. Similarly, a character string 1722 (Classical) is obtained from multidimensional vector 1711. Furthermore, a character string 1724 (Digital Camera) is obtained from multidimensional vector 1713.

[0062] Fig. 18B is a diagram showing an example of information related to a manuscript processed by a vector search server. Corresponding word search result 1806 shown in Fig. 18B(a) includes character string 1807 (SLR), character string 1803 (classical), and character string 1804 (digital camera). File name 1810 shown in Fig. 18B(b) is generated from character strings 1807 to 1809.

[0063] Although preferred embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments and various modifications and variations are possible within the scope of the gist thereof. The present invention can also be realized by supplying a program that realizes one or more functions of the above-described embodiments to a system or device via a network or storage medium, and having one or more processors in the computer of the system or device read and execute the program. The present invention can also be realized by a circuit (e.g., an ASIC) that realizes one or more functions. Furthermore, while the processing unit 307 is built into the generation AI server 300 in the above-described embodiments, this is not limited thereto and may be built into the MFP 101, for example.

[0064] In the image forming system 1000, for example, the generation AI server 300 and the vector search server 400 may be located outside Japan, while the terminal device, the MFP 101, may be located within Japan. Even in this case, files and data can be transmitted from the server to the terminal device, and the terminal device can receive the files and data. Even if the server is located outside Japan, the transmission and reception (transmission and reception) of files and data in this system are performed as a single entity, i.e., without any separate operation by the user of the terminal device. Since the system functions when a terminal device located within Japan receives the files and data, the transmission and reception can be considered to have occurred within Japan. Furthermore, in this system, even if the server is located outside Japan and the terminal device is located within Japan, the terminal device can still perform the main functions of the system, and the effects of those functions can be realized within Japan. For example, even if the server is located outside Japan, if the terminal device constituting the system is located within Japan, the system can be used within Japan using the terminal device. Furthermore, the use of this system may affect the economic interests of, for example, the patent holder.

[0065] The disclosure of each embodiment includes the following configurations, methods, and programs. (Configuration 1) An image forming apparatus having a scanning function for scanning a document, an image acquisition unit that acquires a scanned image of the document by executing the scan function; an input means for inputting the scanned image to a processing unit capable of processing to generate a file name from the scanned image using a learning model; a file name acquisition means for acquiring the file name generated by the processing unit from the processing unit; a file generation unit that generates a file by naming the scanned image with the file name acquired from the file name acquisition unit. (Configuration 2) An information acquisition unit is provided for acquiring auxiliary information for assisting in the generation of the file name when the processing unit generates the file name; 2. The image forming apparatus according to claim 1, wherein the input unit inputs the auxiliary information together with the scanned image to the processing unit. (Configuration 3) The scanned image contains at least one of information of characters, symbols, figures, and photographs; The image forming apparatus according to configuration 2, wherein the auxiliary information includes at least one of the information included in the scanned image, the name of the language used in the file name, information indicating the level of detail in the file name, and information indicating the purpose of the file. (Configuration 4) The input unit inputs instruction information to the processing unit to instruct the generation of the file name together with the scanned image, 4. The image forming apparatus according to any one of configurations 1 to 3, wherein the processing unit generates the file name when the instruction information is input. (Configuration 5) The manuscript is a printed matter having a plurality of pages, 5. The image forming apparatus according to any one of configurations 1 to 4, wherein the image acquisition unit acquires a scanned image of a first page of the printed matter as the scanned image. (Configuration 6) Prior to the generation of the file by the generation means, it is determined whether or not to name the scanned image with the file name acquired from the file name acquisition means; 6. The image forming apparatus according to any one of configurations 1 to 5, wherein the generating unit generates the file when the file name to be given to the scanned image is determined. (Configuration 7) The image forming apparatus is communicably connected to an information processing apparatus that is an external apparatus, The image forming apparatus according to any one of configurations 1 to 6, wherein the processing unit is capable of executing at least one of image recognition processing and character recognition processing, and is built into the information processing apparatus. (Configuration 8) The image forming device according to any one of configurations 1 to 6, characterized in that the processing unit is capable of executing at least one of image recognition processing and character recognition processing and is built into the image forming device. (Method 1) A method for controlling an image forming apparatus having a scanning function for scanning a document, comprising: an image acquisition step of acquiring a scanned image of the document by executing the scan function; an input step of inputting the scanned image to a processing unit capable of processing to generate a file name from the scanned image using a learning model; a file name acquisition step of acquiring the file name generated by the processing unit from the processing unit; a generating step of generating a file in which the scanned image is named with the file name acquired in the file name acquisition step. (Program 1) A program for causing a computer to execute each means of the image forming apparatus according to any one of configurations 1 to 8. [Explanation of symbols]

[0066] 101 MFP 111 CPU 118 Reading unit 126 Communications Department 200 Information terminal 300 Generation AI Server 307 Processing Section 400 Vector Search Server

Claims

1. An image forming apparatus having a scanning function for scanning an original, an image acquisition unit that acquires a scanned image of the document by executing the scan function; an input means for inputting the scanned image to a processing unit capable of processing to generate a file name from the scanned image using a learning model; a file name acquisition means for acquiring the file name generated by the processing unit from the processing unit; a file generation unit that generates a file by naming the scanned image with the file name acquired from the file name acquisition unit.

2. an information acquisition unit that acquires auxiliary information that assists in the generation of the file name when the processing unit generates the file name; 2. The image forming apparatus according to claim 1, wherein the input unit inputs the auxiliary information together with the scanned image to the processing unit.

3. The scanned image includes at least one of information of characters, symbols, figures, and photographs; 3. The image forming apparatus according to claim 2, wherein the auxiliary information includes at least one of the information contained in the scanned image, the name of the language used in the file name, information indicating the level of detail in the file name, and information indicating the purpose of the file.

4. the input means inputs, to the processing unit, instruction information instructing the processing unit to generate the file name together with the scanned image; 2. The image forming apparatus according to claim 1, wherein the processing unit generates the file name when the instruction information is input.

5. the manuscript is a printed matter having a plurality of pages, 2. The image forming apparatus according to claim 1, wherein the image acquisition unit acquires a scanned image of a first page of the printed matter as the scanned image.

6. Prior to the generation of the file by the generation means, it is determined whether or not to name the scanned image with the file name acquired from the file name acquisition means; 2. The image forming apparatus according to claim 1, wherein the generating unit generates the file when the file name to be given to the scanned image is determined.

7. the image forming apparatus is communicably connected to an information processing apparatus which is an external apparatus, 2. The image forming apparatus according to claim 1, wherein the processing unit is capable of executing at least one of image recognition processing and character recognition processing, and is built into the information processing apparatus.

8. 2. The image forming apparatus according to claim 1, wherein the processing unit is capable of executing at least one of image recognition processing and character recognition processing, and is built into the image forming apparatus.

9. A method for controlling an image forming apparatus having a scanning function for scanning an original, comprising: an image acquisition step of acquiring a scanned image of the document by executing the scan function; an input step of inputting the scanned image to a processing unit capable of processing to generate a file name from the scanned image using a learning model; a file name acquisition step of acquiring the file name generated by the processing unit from the processing unit; a generating step of generating a file in which the scanned image is named with the file name acquired in the file name acquisition step.

10. 2. A program for causing a computer to execute each of the means of the image forming apparatus according to claim 1.

Images