GB2404053A - Document scanning and correction terminal - Google Patents

Document scanning and correction terminal Download PDF

Info

Publication number
GB2404053A
GB2404053A GB0415661A GB0415661A GB2404053A GB 2404053 A GB2404053 A GB 2404053A GB 0415661 A GB0415661 A GB 0415661A GB 0415661 A GB0415661 A GB 0415661A GB 2404053 A GB2404053 A GB 2404053A
Authority
GB
United Kingdom
Prior art keywords
document
terminal
customer
user
terminal according
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Withdrawn
Application number
GB0415661A
Other versions
GB0415661D0 (en
Inventor
Nasser Sherkat
Joanne Coy
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Royal Mail Group Ltd
Original Assignee
Royal Mail Group Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Royal Mail Group Ltd filed Critical Royal Mail Group Ltd
Publication of GB0415661D0 publication Critical patent/GB0415661D0/en
Publication of GB2404053A publication Critical patent/GB2404053A/en
Withdrawn legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/10Office automation; Time management
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00Handling natural language data
    • G06F40/10Text processing
    • G06F40/166Editing, e.g. inserting or deleting
    • G06F40/174Form filling; Merging
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00Handling natural language data
    • G06F40/20Natural language analysis
    • G06F40/232Orthographic correction, e.g. spell checking or vowelisation

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Artificial Intelligence (AREA)
  • Strategic Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Human Resources & Organizations (AREA)
  • General Business, Economics & Management (AREA)
  • Data Mining & Analysis (AREA)
  • Tourism & Hospitality (AREA)
  • Quality & Reliability (AREA)
  • Economics (AREA)
  • Operations Research (AREA)
  • Marketing (AREA)
  • Character Discrimination (AREA)

Abstract

A terminal, such as a customer service terminal, comprising: means for automatically analysing a digital document to reveal deficiencies in that document; and means for inputting data to that document for the correction of said deficiencies.

Description

TERMINAL
FIELD OF THE INVENTION
This invention relates to terminals in general, and to customer service terminals in particular. A particularly preferred embodiment of the invention relates to a customer service terminal that is operable to assist customers with document completion.
BACKGROUND TO THE INVENTION
It is often the case that customers seeking to request or purchase a service from a given organization will be asked to complete a document, such as a pre-printed form for example, which specifies details associated with the customer and/or with the service that they are requesting or purchasing.
As an illustrative non-limiting example, a customer who is about to move from one home to another might wish to set up an automated mail forwarding service to ensure - at least for a predetermined period of time 1 S that any mail addressed to their old address is automatically redirected to their new address.
To implement such a service in the United Kingdom, a customer would typically go to their local Post Office and fill in a form to provide various details such as their current address, their new address, the addressees for whom mail is to be forwarded, and the type of service they require (e.g. a one month forwarding service, or a six month forwarding service etc.) The customer would also typically have to provide evidence of some sort to demonstrate that it is indeed their mail that they are seeking to have forwarded.
Once the customer has completed the form, they then would typically have to stand in a queue until one of the Post Office customer service representatives becomes available. Once the customer gets to the counter the representative will check the details provided by the customer on the form, and check the validity of any identification that might be required. If the customer has completed the form correctly, the representative will collect payment and the transaction will be concluded.
If, as is often the case, the customer has made one or more errors or omissions when completing the form, the representative will point this out to the customer and ask them to make the appropriate correction(s). For simple errors, the customer may be able to make the appropriate correction(s) whilst they are standing at the counter. For more serious errors or omissions, however, in order to reduce any inconvenience for other customers waiting in the queue, the customer may be asked to vacate the counter and correct the form. If this should happen then the customer may then have to wait in the queue once more before they can resubmit the form for scrutiny by the representative.
It could even be the case that the customer does not have all the information that they require to complete the form correctly. For example, it is not unusual for someone moving house not to know the full post- code for their new address. If such a customer should attempt to request a mail redirection service, then they may well be told that the form cannot be accepted until they have that new post-code, and the customer may then be faced with having to leave and return another day when they have the information that is required.
An even worse scenario is one where the customer has made mistakes on the form that are not picked up by the representative. In such circumstances the errors may only transpire at a much later date, and the expense associated with tracking down the customer to correct the error can be significant.
For customers who are in employment, it is often the case that the most convenient time available to them for undertaking these administrative tasks is their lunch break. It is not unusual, therefore, for large numbers of customers to descend on Post Offices at or around lunchtime. If, as is sometimes the case, a number of customer service representatives are also on their lunch break at this time, then large queues can quickly build up. It will be appreciated that in such circumstances it only requires a few customers in the queue to be undertaking administrative tasks of the type described above for serious delays to result.
Another important factor to consider is that in today's multi-cultural society not all of the attending customers will have English as their mother tongue, and for such customers the prospect of having to complete a relatively complicated administrative document in a foreign language can be quite daunting. The representatives can also find it difficult and frustrating to explain inadequacies in submitted forms to customers who cannot understand the majority of what is being said to them.
In an effort to alleviate this problem, it has previously been proposed to print these administrative forms in a number of different languages, but this provides no assistance to a representative who has to explain deficiencies in a submitted form to the customer. Furthermore, the costs associated with having many forms translated into a number of different languages, and then reprinted can be quite significant, as can the amount of space required to display multiple copies of the various forms.
Such problems are not limited solely to those customers who do not have English as a mother tongue. Disabled and elderly customers, for example, can find it very difficult to read and properly complete these administrative documents.
Another problem facing the aforementioned customer service representatives is that it is become commonplace for the efficiency or productivity of a given organisation to be adjudged by the time taken for a given customer to be served. Clearly, if the organization has a relatively high proportion of customers who are undertaking administrative tasks of the kind described, then the likelihood of delays occurring are quite significant, and this can have a negative impact on the performance assessment of that organization and, by inference, it's representatives.
One might think that employing a greater number of customer service representatives could simply solve this particular problem. However, it is often the case that there simply is not enough floor-space in the building to create any further customer service points. It is also the case that employing farther representatives would increase the cost base of the organization, and this would affect the profitability of the organisation.
It is apparent, therefore, that there are a number of significant problems associated with the completion of documentation, for example administrative documentation, by customers. These problems are faced by many organizations, of which the Post Office is merely one illustrative
example.
OBJECT AND STATEMENT OF THE INVENTION
It is an aim of the present invention to mitigate at least some of the problems described above, and to this end one presently preferred embodiment of the invention provides a terminal, such as a customer service terminal, comprising: means for automatically analysing a digital document to reveal deficiencies in that document; and means for inputting data to that document for the correction of said deficiencies.
A second aspect of the invention relates to a computer program for use in a (customer service) terminal, the computer program comprising one or more software portions which, when executed in an execution environment, are operable to: automatically analyse a digital document to reveal deficiencies in that document; and automatically input data to that document for the correction of one or more of said deficiencies.
Preferred features of each of these aspects of the invention are set out in the accompanying claims.
BRIEF DESCRIPTION OF THE DRAWrNGS
Fig. 1 is a schematic depiction of a terminal according to a preferred embodiment of the invention; Fig. 2 is a schematic representation of the various processes undertaken in the terminal of the preferred embodiment; and Fig. 3 is a more detailed schematic representation of the processes shown in Fig. 2.
DETArr ED DESCRIPTION OF PREFERRED EMBODIMENTS
Preferred embodiments of the invention will now be described in detail with particular reference to a customer service terminal for use in assisting the completion of forms in a Post Office. It should be noted, however, that the present invention is not limited to this particular use or indeed to assisting with forms, and may instead be used in a number of different organizations to assist with the completion of many different documents. As a consequence, the following description should be read as being illustrative rather than restrictive.
Fig. 1 is an illustrative representation of a customer service terminal.
As shown, the terminal comprises a housing 1 in which a plurality of individual components are provided. The components are interconnected by means of a system bus 3, and the system bus is operable to pass data and/or instructions between the various internal and/or external components of the terminal 1. As shown, the terminal architecture is generally modular, and this allows components to be easily replaced or upgraded as required throughout the life of the terminal.
Connected to the bus is a processor 5 which is operable to execute software that controls the terminal. A memory 7 and data storage device 9 (such as a hard disk) are also provided for software and/or data storage.
The terminal is provided with a document scanner 11 and a printer 12, each of which is connected to the system bus 3. The document scanner 11 is operable to scan a physical document, such as a form for example, to provide a digital version of that document for subsequent processing. The scanner may be of the so-called "flatbed" type, or alternatively may instead comprise a tractor feed which is operable to draw a document through the scanner. The latter is preferable as it allows most of the components of the scanner to be hidden inside the terminal housing 1. In use, customers would simply feed documents through a slot in the housing. A flatbed scanner, whilst conceivable, would be unlikely to stand up to the rigours of public use. The printer may conveniently be of any type such as a laser, ink jet or dot matrix printer.
In a highly preferred arrangement the document scanner and printer are integrated so that a partly completed document can be inserted into the terminal, data can be printed on to the document to complete it and the completed document can then be output from the terminal.
The system bus 1 is also connected to "input means" 11 and "output means" 13. "Input means" in the context of the present application is intended to encompass any and all devices (without restriction) that can be operated by a customer to input instructions and/or data into the terminal.
Similarly, "output means" is intended to encompass any and all devices (again without restriction) which are operable to provide instructions or data to a customer. The input means may comprise, for example, a microphone, keyboard and mouse (as shown in Fig. 1). The output means may comprise, for example, one or more loudspeakers and a display.
As a further example, the terminal may comprise a display with a touchsensitive overlay (a so-called touch-sensitive screen). With such a device, data and/or instructions could be displayed to the customer and the customer could enter data or select options simply by pressing the appropriate point on the screen.
In another preferred arrangement, the terminal may be voice activated and to implement this, the processor may be configured to run appropriate voice recognition software that is arranged to interpret sound data received from the microphone. To be accessible for all users, such a voice-operated terminal would probably additionally comprise a keyboard or a touch sensitive screen.
Advantageously, voice recognition may be implemented using finite state networks. By this we mean that the algorithm implemented by the processor when interpreting input sound data is configured to make use of a Smite lexicon of words that might normally be associated with the completion of various documents.
As shown in Fig. 1, in the preferred embodiment, the system bus 3 of the terminal 1 is coupled to a communications interface 15, and via the interface 15 to an input/output port (LO port) 17. The interface 15 and port 17 are provided to enable the terminal to communicate, for example via an internet, local area network or a dedicated network, with one or more external components such as a customer service centre 19 and/or a database 21. In a highly preferred embodiment, the interface and port are configured to permit a live video link to be established between the terminal and a customer representative at the customer service centre. In this arrangement, the terminal may also comprise a video camera (not shown) so that the customer representative is provided with an image of the customer. Such an arrangement would allow the representative to "speak" with customers using a non-verbal language such as sign language; sign-language symbols made by the customer representative being displayed to the customer on the display.
The database may include data that can be used to assist in the completion of a given partly completed document. For example, the database could include details of appropriate postcodes for addresses in the United Kingdom. The database could conceivably be included within the terminal, and such an architecture would help speed operation of the terminal.
However, with such an arrangement it would then be difficult to ensure that each terminal is properly updated, and for this reason it is preferred for the database to be held centrally where it can be interrogated by a number of terminals and easily updated as required.
Fig. 2 is a schematic representation of the various processes undertaken in the terminal of the preferred embodiment. Most (if not all) of these processes will be implemented by means of software, although it is conceivable of course that similar functionality may be provided in hardware, for example by means of application specific integrated circuits (ASICs).
The terminal will, before use, be in a stand-by state awaiting wake-up by a customer. Wake-up may occur, for example in response to a key or other input device being pressed or operated. A message may be displayed to customers informing them that a key (or other input device) must be pressed to proceed.
Once woken, the terminal will prompt the customer- for example by means of the screen and/or a message relayed to the customer via the loudspeaker to input the document (in this example a form) that they have selected. In most cases it is envisaged that the document will be partially completed. The terminal will, however, be able to assist in the completion of a new document which the customer has not started completing. The terminal will also be operable to assist with the preparation of a digital document that is selected by the customer from a list. In this event, only once the document has been completed is it printed for the customer.
Document input is sensed by means of an appropriate sensor that detects when a document has been inserted into the terminal. Once the insertion of a document is detected, the tractor feed (in the preferred embodiment) of the scanner is operated to draw the document into the terminal (step 25) and through the scanner. As the document is drawn through the scanner an optical reader scans the document (step 27) and digital data representing the document is generated. This digital data set will hereafter be referred to as a digital document 29.
In the case of a form, the digital document will comprise two discrete data sets. The first data set pertains to pre-printed parts of the form that the terminal can effectively ignore. The second data set pertains to any data written, printed or typed on the form by the customer, and may (as explained above) comprise no data at all.
Once the digital document has been scanned it is compared with a series of stored templates each comprising a digital document of a blank form.
If the first data set of the scanned document matches a said template, the processor determines that the customer has been trying to complete a particular form and loads the appropriate software for completion of that form (step31).
The processor then invokes optical character recognition (OCR) software to interpret data of the second data set (step 33), if any, and to determine whether any parts of the form have yet to be completed (step 35) .
The processor may then opt to consult the aforementioned remote data base (step 37) in the event that any missing information can be inputted to the digital document from the data stored in the database. For example, if the customer has written their new address on the form but has failed to include a postcode, the processor may (based on the address determined from the foregoing OCR step) retrieve the correct postal code from the database and insert that postal code at an appropriate point in the digital document.
In the event that the processor is not able to retrieve data to complete the document, a message is relayed to the customer prompting them to input the necessary data (step 39). That message may comprise a displayed textual message. Additionally, or alternatively, the message may be accompanied by an appropriate audio message relayed to the customer via the aforementioned loudspeaker.
a highly preferred embodiment, the customer will be able to select a language from a range of options, and the processor will be configured to display textual messages and/or relay audio messages in that language.
Once the request for information has been relayed to the customer, it is then up to the customer to operate the input means to provide the requested data to the terminal (step 41). Data input by the customer to the terminal (step 43) is then included in an appropriate point in the second data set of the digital document. This process is then repeated until all necessary information has been inputted into the digital document. Once the document is complete it is passed to the printer for printing (step 45).
The printing step may comprise the printing of additional material onto a document which has been partially completed by the customer.
Alternatively, the printing step could instead comprise the printing of the entire digital document. Once the document has been completed the customer can take it to a customer service point for further processing.
In the event that the customer should choose not to input a partially completed form and should instead opt to select a form from a list of options, it will be apparent that steps 33 and 35 (and possibly also step 37) will not then be required.
If the customer should experience problems, or prefer to deal face-to face with a representative, then the customer can initiate (step 47) a live video link with a human representative at a customer service centre. The customer representative is preferably able to control the terminal remotely from the service centre, and this permits the representative to complete the document rather than simply providing instructions to the customer for completion of the document.
In summary, this embodiment of the invention pertains to the general concept of a terminal which permits the scanning of documents, the identification of missing or erroneous data (for example by reference to a predefined template) in those documents, the input of any missing or erroneous data (by the customer and/or from a remote database), and the printing of a completed and/or corrected electronic document comprised of the data originally submitted (if any) and the inputted data.
Fig. 3 is a more detailed schematic representation of the processes shown in Fig. 2. This aspect of the preferred embodiment will now be described with particular illustrative reference to the completion of a mail redirection form which includes the following fields: customer's existing address, customer's new address, person(s) for whom mail is to be forwarded, and duration of mail forwarding service. In this example we will assume that the customer has only handwritten their full existing address and partial new address on the form - although it should of course be remembered that the customer could opt to complete the entire form by means of the terminal.
As described above, the terminal on wake-up replays a message, for example an audio or video message, to the customer which instructs the customer to insert their form into the terminal (step 50). The customer may be asked at this point to select a language in which instructions and messages will be provided.
Following insertion of the form it is drawn into the terminal by the aforementioned tractor feed, and scanned (step 52). Scanning of the form generates a digital image of that form which is then processed to remove any unwanted artefacts or other distortions (step 54).
The scanned and processed form is then compared with a series of form templates held in an electronic data storage device. If the form presented should match, to a predefined extent, one of the aforementioned templates, the processor sets a flag indicating that the form presented is of a particular type (step 56), and retrieves data applicable to that particular form type. For example, the processor could determine that the form presented is a mail forwarding form simply by counting the number of areas provided on the form for customer data. The form could be constructed to assist this process, for example by surrounding areas on the form areas with a marker that indicates that the area inside the marker is for customer data.
The processor then performs a contextual analysis of the various fields contained within the form by reference to a database (step 58), before analysing any parts of the form which have been completed by the customer to determine whether the customer has chosen to hand-write or print data onto the form (step 60). Once the processor has determined whether the form has been handwritten or printed, a flag is set (step 62) and the appropriate optical character recognition software is loaded (step 64). Data on the form is then deciphered by the OCR software (step 66) with reference to a lexicon 68. The deciphered text is stored as a digital form 70.
The processor then determines whether the form includes any omissions or errors (step 72) that can be completed or corrected automatically. For example, the processor might correct any misspellings, or incorrect details such as birth dates which do not tally with a stated age.
In this particular example, the processor would determine that the deciphered "new address details" did not include a post code, and would refer to a database 74 for retrieval of the correct post code for the address written or printed by the customer. Once the correct post code has been retrieved it is written to the digital form 70 (step 76).
The processor would then compare the digital form 70 with the aforementioned template and determine that the fields "person(s) for whom mail is to be forwarded" and "duration of mail forwarding service" had not I been completed (step 78). As these particular data items cannot reliably be completed without reference to the customer, the processor relays a message to the customer prompting them to input the necessary information (step 80).
The message might comprise, for example, a textual message displayed on the display. Alternatively, or additionally, the message may comprise a synthesised audio message which can be played to the customer via the loudspeaker. In the preferred embodiment, the message is provided to the customer in the language they have previously chosen.
The customer can react to the message by inputting the missing information, and this could be accomplished - for example - simply by speaking into the microphone. A speech engine 82, under control of the processor, would interpret spoken text with the aid of the lexicon to derive the missing information and that information would then be added to the digital form 70. In the preferred embodiment, the text deciphered by the speech engine would be displayed to the customer for checking before being stored in the digital form 70.
In this instance it is envisaged that the processor will control the loudspeaker to play a message to the customer which advises that no addressee information has been entered on the form, and requests the customer to input that information. Once this information has been received and processed, the processor would then play a second message asking how long the customer wishes the mail forwarding service to be in operation.
Various fall-back procedures will also function to allow for proper operation, for example in circumstances where a customer has not spoken with sufficient volume or their speech cannot clearly be interpreted. As an alternative to speech recognition, the customer may instead be prompted to use a keyboard to enter the missing data, or to use a touch sensitive screen.
Once the data has been inputted and optionally checked by the customer, it is added to the digital form to complete it. Once completed the processor controls the printer (step 84) to print the new data inputted by the f customer and the missing data retrieved from the database onto the form inserted by the customer into the terminal. The printed form is then output from the terminal, and the customer can take the form to a customer service point for further processing. In a highly preferred embodiment, the processor will also provide the customer with a message advising them of any other requirements that must be met before the form can be accepted. For example, in this example, the processor might display a message advising the customer that they will need appropriate identification before the request can be accepted, and further that all addressees for whom mail is to be forwarded must have signed the form. An advantage of this is that a customer who does not have the appropriate identification or has not had the form signed by all the addressees, will know that they require further documents without having had to queue to speak with a customer service representative.
If the customer should experience problems, or prefer to deal face-to face with a representative, then the customer can initiate (step 86) a live video link with a human representative at a customer service centre. The customer representative is preferably able to control the terminal remotely from the service centre, and this permits the representative to complete the document rather than simply providing instructions to the customer for completion of the document.
In another preferred embodiment of the invention, the terminal may be located within a booth. The booth providing privacy and security to the user of the terminal, as well as providing controlled ambient conditions such as light and background noise levels. A seat may be provided for customer use, and may be hinged to one wall of the booth so that it can be pivoted out of the way to provide easy access to the terminal for wheelchair users.
In this arrangement is it proposed to provide a dual screen display that the user will face when stood or seated in front of it. The first screen is intended to be primarily for display of information and messages to the user, and as such is mounted generally vertically on a wall of the booth (probably at a slight angle to the vertical to facilitate viewing of information on the screen). This first display can also provide a live video image of an operator in circumstances where video conferencing with the user is necessary or expedient. In the preferred arrangement the first display includes a touch sensitive overlay to allow users to select options simply by touching the screen.
The second screen is mounted on the same wall of the booth at a position below the first screen. The second screen extends generally horizontally from the wall of the booth and is intended to function both as a means to display information to the user, and an ergonomically easy surface for theuser to write on. The second display may be used, for example, for displaying document images (be they electronic originals, or images of paper document scanned by the scanner), and is provided with a highresolution touch-sensitive overlay so that pen or stylus traces can be sensed. By virtue of this arrangement, the user can enter information, correct entries on the form image or sign the form simply by writing on the display. Preferably the second display will be large enough to display a page of A4 in a 1:1 scale without the need of scrolling or zooming.
It is also proposed to provide a dual camera system that is operable to detect the presence and actions of the user. In the envisaged terminal, a first camera will be used for detecting the presence of the user and automatically tilting, panning and zooming on the user's face, and a second camera will used as a means for tracking movements of the user's eyes so that the terminal can determine (with the aid of appropriate software) where the user is looking, and from this detemmination predict the user's likely next actions. In the envisaged arrangement, the cameras will be physically located above the screen but hidden from the user in order to avoid distraction.
In effect, the first camera provides (by providing a macro view for detecting the face and movements of the user) the context for the second eye gaze tracking camera. This first camera can also be used by an operator to see the user, such functionality being required for video conferencing with a user or for holding a sign language conversation with a user.
As aforementioned, the terminal will include a set of loudspeakers (probably one on either side of the vertical display) through which information can be conveyed to the user. These messages can be in the form of a synthesised voice, or can comprise live speech from an operator engaged in a video conference with the user. The terminal will also include a microphone for user voice input. In the preferred arrangement the terminal will also include an induction loop that may be used by the hearing impaired to facilitate operation of the terminal.
A document scanner mounted in the wall of the booth - for example between the first and second screens - is provided to enable documents to be scanned electronically by the terminal. As aforementioned, in the envisaged arrangement the scanner has a feed mechanism that is operable to enable the scanning of booklets as well as single-sided and double- sided single sheet forms. In the preferred arrangement, the scanner is a colour scanner with colour drop-out filters for extraction of filled fields in any scanned form.
The envisaged terminal also includes the necessary hardware to enable connection over a network with remote operators or remote data storage facilities. This functionality could be accomplished, for example, by providing a means for the terminal to connect to an intranet (i.e. a private network) for the purpose of transferring information such as filled forms or holding video conferencing between operators and users.
If desired, the terminal may also include a finger print scanner so that fingerprints a user of may be compared with fingerprint scans stored in a remote database, to thereby validate the identity of that user. Alternatively, or additionally, the terminal may include a software module that enables iris recognition - for example by comparison of an image captured from the aforementioned eye-gaze tracking camera, with an image stored in a remote database.
To enable the customer to pay for any requested service, the temminal also preferably comprises a means for accepting payment - such as a credit or debit card scanner or a cash receiving device (the like of each of which are well known in the art and hence will not be described in detail herein).
Local storage will be provided, for example for storing the software under whose control the terminal operates. That storage may comprise, for example, one or more hard disks, EPROM and RAM.
Operation of the terminal will be controlled by a central processing unit (CPU), and in general terms the CPU will be used to execute all local tasks. As illustrative examples, the CPU may execute software to perform tasks such as management and synchronization of peripheral processing units (for example for eye-gaze tracking, hand writing recognition and speech recognition), as well as software that controls the user interface and connectivity with an external network. If the CPU has sufficient processing power, then all of these functions may be accomplished by one processing unit. In the alternative, the CPU may instead comprise a number of individual processing units that together provide a multiprocessor platform.
As mentioned above, the functionality provided by the terminal is controlled by software executed by the CPU. In an envisaged arrangement the terminal will include a number of software modules each for accomplishing designated tasks. In addition to normal tasks such as communications control, some of these modules will accomplish more complicated tasks - the like of some of which are briefly mentioned below.
For example, implement document recognition the terminal may include preproc essing, dynamic thresholding and/or co tour quantisati on software modules that are operable, for example, to pre-process scanned document images to re-align, de-skew or rotate images derived from scanned documents.
The terminal will also include a document image layout analysis module that is operable to achieve functionality such as field extraction, and pre-classification of field contents into machine printed, hand printed or unconstrained handwriting. Once fields have been classified, other modules may be invoked as required to provide functionality such as optical character recognition, hand printed character recognition, unrestrained cursive script recognition, and signature verification.
To implement the functionality described above in connection with the dual camera system, the terminal may include a face recognition module that is operable to locate the face of a user, and an eye-gaze tracking module that is operable to track the eyes of the user to thereby determine where the user is looking. The face recognition module may also be able to track movement of a user's lips (and potentially to interpret such movement by reference to a lexicon for user input), and/or facial impressions.
If fingerprint or iris recognition functionality is provided, then the terminal will comprise appropriate software modules to enable the appropriate images to be captured and compared with images stored remotely.
To facilitate interaction with a user, the terminal may include software modules that are operable to recognise gestures made by a user, to produce synthesised speech from text, to recognise user speech, and to process user hand writing. Such modules may be controlled by an intelligent software agent that is operable to symbiotically combine a plurality of software modules to mimic, at least as far as is possible, a natural human interaction with the user.
Whilst various preferred embodiments of the present invention have been described above in detail, it will be apparent to those persons skilled in the art that modifications may be made to the particular embodiments disclosed without departing from the spirit and scope of the invention.
For example, whilst the foregoing description has referred to touch sensitive screens as comprising touch-sensitive overlays, it will be immediately apparent to those persons skilled in the art that a touch sensor can be integrated into the screen as opposed to being overlaid on it. It will also be apparent that touch sensitivity can be achieved through a variety of technical means such as magnetic or capacitive measurement.
As regards the software used to implement the above described functionality, it will be appreciated by those of skill in the art that any of a number of different programming languages may be employed. The applicant's preferred current option is to use an object-orientated programming language, but the functionality described herein may be coded in any other programming language (such as procedural language for
example).
As a further example of one possible modification, that could enhance interactive sessions between customers and terminals, an avatar could be animated as textual or audio messages are replayed to the customers.
A last point of note is that whilst particular combinations of features herein described have been set out in the accompanying claims, the scope of the present invention extends to any combination or permutation of features described herein irrespective of whether or not that particular combination or permutation has been explicitly enumerated in the accompanying claims.

Claims (27)

1. A terminal, such as a customer service terminal, comprising: means for automatically analysing a digital document to reveal deficiencies in that document, and means for inputting data to that document for the correction of said deficiencies.
2. A terminal according to Claim 1, comprising means operable to generate said digital document from a real physical document.
3. . A terminal according to Claim 1, wherein said generating means comprises a scanner configured to scan a real physical document presented by a customer.
4. A terminal according to Claim 1, wherein the digital document is selected by a customer from a plurality of stored digital documents.
5. A terminal according to any preceding claim, comprising means for printing a corrected document comprised of said digital document and any inputted data.
6. A terminal according to Claim 5 when dependent on Claim 3, wherein the printing means is operable to print onto the real physical document presented by the customer in order to provide said corrected document.
7. A terminal according to any preceding claim, wherein the analysing means comprises: means for comparing the digital document with a template to thereby identify one or more areas of the document which require the insertion of data for completion.
8. A terminal according to Claim 7, comprising means for determining that an omission of data has occurred in the event that one or more of said areas is without data.
9. A terminal according to any preceding claim, comprising means for determining textual information from the digital document, and means for automatically correcting any likely errors in that textual information.
10. A terminal according to Claim 8 or 9, comprising means for automatically retrieving, from a data store, likely data for correction of said omission or said error.
11. A terminal according to Claim 8 or 9, comprising means for outputting a message to a customer prompting them to input data for the correction of said omission or said error. 15,2
12. A terminal according to Claim 11, wherein said output means is operable to present a visual and/or and audible message to the customer.
13. A terminal according to Claim 10 or 11, wherein said output means comprises one or more of a loudspeaker and a visual display unit.
14. A terminal according to any preceding claim, wherein said input means comprises one or more of a microphone, a keyboard or a pointing device (e. g. a mouse).
15. A terminal according to Claim 14, wherein the input means includes a microphone, and the terminal includes means operable to permit voice control of the terminal.
16. A terminal according to Claim 15, wherein the voice control means is implemented by means of finite state networks.
17. A terminal according to Claim 16, wherein said finite state networks include one or more algorithms each of which is configured to compute within a finite set of values.
18. A terminal according to Claim 16 or 17, comprising a lexicon of likely terms for use with said voice control.
19. A terminal according to any preceding claim, comprising means operable to establish a live video link between the terminal and a customer service representative usually, but not always, located remotely from the terminal.
20. A computer program for use in a customer service terminal, the computer program comprising one or more software portions which, when executed in an execution environment, are operable to: automatically analyse a digital document to reveal deficiencies in that document; and automatically input data to that document for the correction of one or more of said deficiencies.
21. A computer program according to Claim 20, comprising one or more software portions that are configured to implement customer voice control by means of finite state networks.
22. A computer program according to Claim 21, comprising one or more software portions operable convert customer speech to data for inclusion in said document for the correction of one or more of said deficiencies.
23. A computer program according to any of Claims 20 to 22, comprising one or more software portions configured to automatically retrieve data from a database for the correction of one or more of said deficiencies.
24. A customer service terminal comprising means for scanning a paper document to generate a digital copy document, means operable to analyse the digital copy document to identify deficiencies, means operable to generate corrections for any deficiencies that are automatically correctable without reference to a user, means to solicit corrections from a user for correction of any deficiencies that are not automatically correctable, and means for applying any generated and solicited corrections to the scanned paper document.
25. A customer service terminal comprising: a scanner that is operable to scan a document input to the terminal by a user to generate a digital version thereof; a computer; a document analysis software module executable by the computer to identify deficiencies in the digital version of said scanned document; an information gathering software module executable by said computer to automatically acquire information for correcting any deficiencies identified in said digital version that may be correctable without reverting to the user, and to generate a user prompt for the supply of information for correction of any identified deficiencies that cannot be corrected without reverting to the user; apparatus for providing said user prompt to the user; a user input device operable by the user in response to the provision of a user prompt to input information to the computer; a printer operable to correct at least some of said identified deficiencies by printing on the document inputted by the user any information automatically acquired by the information gathering software module and any information input by the user; and a document feeder that is operable to transfer the document from the printer and output the document for return to the user.
26. A terminal substantially as hereinbefore described with reference to the accompanying drawings.
27. A computer program substantially as hereinbefore described.
GB0415661A 2003-07-16 2004-07-13 Document scanning and correction terminal Withdrawn GB2404053A (en)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
GBGB0316650.1A GB0316650D0 (en) 2003-07-16 2003-07-16 Terminal

Publications (2)

Publication Number Publication Date
GB0415661D0 GB0415661D0 (en) 2004-08-18
GB2404053A true GB2404053A (en) 2005-01-19

Family

ID=27763928

Family Applications (2)

Application Number Title Priority Date Filing Date
GBGB0316650.1A Ceased GB0316650D0 (en) 2003-07-16 2003-07-16 Terminal
GB0415661A Withdrawn GB2404053A (en) 2003-07-16 2004-07-13 Document scanning and correction terminal

Family Applications Before (1)

Application Number Title Priority Date Filing Date
GBGB0316650.1A Ceased GB0316650D0 (en) 2003-07-16 2003-07-16 Terminal

Country Status (1)

Country Link
GB (2) GB0316650D0 (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8413037B2 (en) 2010-06-27 2013-04-02 Hewlett-Packard Development Company, L.P. User selection of flaw present within digitally scanned document
US9183536B1 (en) * 2010-10-22 2015-11-10 United Services Automobile Association (Usaa) Service representative and remote location document communication

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4573187A (en) * 1981-07-24 1986-02-25 Asulab S.A. Speech-controlled electronic apparatus
US20030020963A1 (en) * 2001-07-25 2003-01-30 Currans Kevin G. Data acquisition system and method using answer forms
WO2003094073A1 (en) * 2002-04-30 2003-11-13 Medco Health Solutions, Inc. Prescription management system
EP1410311A1 (en) * 2000-03-24 2004-04-21 Siemens Dematic Postal Automation L.P. Method and system for form processing

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4573187A (en) * 1981-07-24 1986-02-25 Asulab S.A. Speech-controlled electronic apparatus
EP1410311A1 (en) * 2000-03-24 2004-04-21 Siemens Dematic Postal Automation L.P. Method and system for form processing
US20030020963A1 (en) * 2001-07-25 2003-01-30 Currans Kevin G. Data acquisition system and method using answer forms
WO2003094073A1 (en) * 2002-04-30 2003-11-13 Medco Health Solutions, Inc. Prescription management system

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8413037B2 (en) 2010-06-27 2013-04-02 Hewlett-Packard Development Company, L.P. User selection of flaw present within digitally scanned document
US9183536B1 (en) * 2010-10-22 2015-11-10 United Services Automobile Association (Usaa) Service representative and remote location document communication
US9390421B1 (en) 2010-10-22 2016-07-12 United Services Automobile Association (Usaa) Service representative and remote location document communication
US9741041B1 (en) 2010-10-22 2017-08-22 United Services Automobile Association (Usaa) Service representative and remote location document communication
US10521803B1 (en) 2010-10-22 2019-12-31 United Services Automobile Association (Usaa) Service representative and remote location document communication
US10949862B1 (en) 2010-10-22 2021-03-16 United Services Automobile Association (Usaa) Service representative and remote location document communication

Also Published As

Publication number Publication date
GB0316650D0 (en) 2003-08-20
GB0415661D0 (en) 2004-08-18

Similar Documents

Publication Publication Date Title
US11310223B2 (en) Identity authentication method and apparatus
US10133385B1 (en) Transaction automation and archival system using electronic contract disclosure units
US9582484B2 (en) Methods and systems for filling forms
US8286083B2 (en) Copying documents from electronic displays
US20140036099A1 (en) Automated Scanning
US8482827B2 (en) Annotation on media sheet indicating functionality to be performed in relation to image on media sheet
CN110322643A (en) Intelligent government affairs services system and its application
US20050251016A1 (en) Image acquiring device and method or rotating images
WO2011126515A1 (en) Authenticating a person's identity using rfid card, biometric signature recognition and facial recognition
US20060285748A1 (en) Document processing device
US9934212B2 (en) Methods and systems for capturing, sharing, and printing annotations
US11025792B2 (en) Image processing apparatus and non-transitory computer readable medium for document processing
CN106256119B (en) Detect the signature line in electronic document
GB2404053A (en) Document scanning and correction terminal
JP2020021458A (en) Information processing apparatus, information processing method, and information processing system
KR102343851B1 (en) Intelligent Civil Service Processing System
KR20090001514A (en) System and method for automating identification card processing task using order waiting system
CN111310588A (en) Online account opening application processing method and device, equipment and medium for gas system
US10776617B2 (en) Sign-language automated teller machine
JP7283245B2 (en) image forming device
US6526184B2 (en) Error-tolerant image computer terminal
WO2023120221A1 (en) Authentication device, authentication system, authentication method, and recording medium
JP2023109651A (en) Program, system, terminal device, method, and information processing apparatus
JP2020106775A (en) Image forming apparatus
JP2023030517A (en) Information processing apparatus, and method for setting information processing apparatus

Legal Events

Date Code Title Description
WAP Application withdrawn, taken to be withdrawn or refused ** after publication under section 16(1)