High volume document image archive system and method

• US 6,115,509 A
• Filed: 03/10/1994
• Issued: 09/05/2000
• Est. Priority Date: 03/10/1994
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Yeskel, Filip Jay
• Primary Examiner(s): BELLA, MATTHEW C
• Application Number:
  US08/209,096
• Time in Patent Office: 2,371 Days
• Field of Search: 382/137, 382/309, 382/311, 382/7, 382/57, 707/1, 707/102, 707/104, 707/530, 345/507, 345/203, 711/100, 364/419.19, 364/552, 395/145, 395/400, 395/600, 358/403, 235/379, 235/479
• US Class Current: 382/309
• CPC Class Codes: G06F 16/40 of multimedia data, e.g. slideshows comprising image and additional audio data retrieval of still image data G06F16/50; retrieval of audio data G06F16/60; retrieval of video data G06F16/70
• CPC Class Codes: G06Q 40/02 Banking, e.g. interest calculation or account maintenance credit or loans G06Q40/03
• CPC Class Codes: G06V 10/987 with the intervention of an operator
• CPC Class Codes: G06V 10/993 Evaluation of the quality of the acquired pattern
• CPC Class Codes: H04N 1/00002 Diagnosis, testing or measuring; Detecting, analysing or monitoring not otherwise provided for error detection, error correction or monitoring in digital computers or digital computer components G06F11/00
• CPC Class Codes: H04N 1/00005 relating to image data
• CPC Class Codes: H04N 1/00039 Analysis, i.e. separating and studying components of a greater whole
• CPC Class Codes: H04N 1/0005 in service, i.e. during normal operation
• CPC Class Codes: H04N 1/00063 using at least a part of the apparatus itself, e.g. self-testing
• CPC Class Codes: H04N 1/00082 Adjusting or controlling interrupting an operation H04N1/00915; inhibiting an operation H04N1/00925
• CPC Class Codes: Y10S 707/99931 Database or file accessing

Abstract

High speed machine scanning of documents such as checks produces digital check images that are placed in archival storage on mass storage devices for later retrieval. Images and/or documents are automatically reviewed by a machine in order to identify images and/or documents that are of suspect quality. Machine review of suspect images and/or documents provides a reject or accept decision. Only acceptable documents are archived. Accepted documents are formed into large data groups that contain a storage location identification for each individual document within the large data group. An index is stored for each such data group wherein the storage location of each document within the large data group is contained. Digital images are selectively converted to visual images, and these visual images are then reviewed by a human operator. This operator review is used to adjust the machine'"'"'s accept/reject decision making process, thereby teaching the machine the correct manner of making its accept/reject decision.

Claims

1. A machine-based document archival method for processing one or more digital images of each of a plurality of documents and digital data that is associated with said documents, comprising the steps of;

machine scanning a document and forming one or more digital images and associated data that correspond to said document,

machine detecting anomalous conditions that may occur during said scanning and/or during subsequent image processing,

defining a plurality of image quality parameters,

machine computing a suspiciousness value for each of said plurality of digital images as a function of said defined image quality parameters and said detected anomalous conditions,

continuing said machine scanning of said plurality of documents independent of the results of said machine computation of suspiciousness value,

making a machine recommendation to archive based upon said machine computed suspiciousness values for each of said plurality of digital images,

machine archive storing said one or more digital images and said associated data in a digital storage device only when a recommendation to archive has been made,

providing machine temporary storage of digital images and associated data prior to said machine archive storing step,

machine erasing said temporary storage after said machine archival storing step,

machine converting into visual images the digital images of a document having one or more digital images that are of suspect quality,

providing human visual review of said visual digital images,

making a human document accept/reject decision based upon said human visual review, and

changing said image quality parameters in a manner to produce future correspondence between said machine computation of suspiciousness value and said human document accept/reject decision.

2. The method of claim 1 including the step of;

assembling a plurality of images and associated data of recommend-to-archive documents into a data group for machine archival storage, said data group containing an index identifying the storage location within said data group of each of the recommend-to-archive documents assembled thereinto, and

said step of machine archive storage including the step of machine identifying the storage location of each data group stored in said digital storage device.

3. The method of claim 2 including the step of;

machine retrieving a selected document from said digital storage device by defining the storage location of the data group containing said selected document,

machine fetching said data group index from said digital storage device at said defined data group storage location,

machine reading the stortage location of said selected document from said fetched index, and

machine fetching said selected document from said digital storage device.

4. A document image processing method, comprising the steps of;

machine scanning a document and forming a plurality of digital images that correspond to said document,

detecting anomalous conditions that may occur relating to said machine scanning, defining a plurality of digital image quality parameters,

machine calculating an image suspiciousness value for each of said plurality of digital images based upon said parameters and said detected anomalous conditions,

machine calculating a document suspiciousness value based upon said parameters and said calculated image suspiciousness values,

using a machine to make a document accept/reject decision based upon said document suspiciousness value,

repeating said method steps to thereby continue said scanning of documents independent of the results of said document accept/reject decisions for each scanned document,

when a document-relect decision has been made, providing machine rescan of said rejected document,

using a machine to assemble a plurality of digital images from a plurality of documents for which a document-accept decision has been made into a data group that contains an index defining the storage location within said data group of said plurality of digital images,

storing said data group in a digital storage device, said storing step including identifying the storage location of each data group that is stored in said digital storage device,

using a machine to convert the digital images corresponding to a document for which a document-relect decision has been made into visual images,

providing review of said visual images by a human operator, and

changing said digital image quality parameters so as to produce future correspondence between said machine accept/reject decision step and said operator visual review.

5. The method of claim 4 including the step of;

using a machine to retrieve at least one digital image of a selected document from said digital storage device by defining the storage location of the data group containing said selected document,

using a machine to fetch said index of said data group from said digital storage device,

using a machine to read the storage location of said selected document from said fetched index, and

using said read storage location to machine fetch said at least one digital image of said selected document from said digital storage device.

6. The method of claim 4 including the steps of;

machine scanning a large plurality of documents to form for each of said documents at least one digital image,

forming a similar large plurality of associated data individually corresponding to each one of said large plurality of scanned documents,

using a machine to assemble the plurality of digital images and the associated data that correspond to a small plurality of documents into a data group,

providing a library level index pointing to the storage location of said data group,

providing a group level index pointing to the storage location of said digital images and associated data in said data group, and

archive storing said library level index, said group level index, said digital images, and said associated data of said data group.

7. The method of claim 6 including the step of;

for each document in said data group, providing a document level index pointing to the storage locations of said digital images and associated data that correspond to each said document, and

storing said document level index.