Image reading apparatus

• US 20050238205A1
• Filed: 03/02/2005
• Published: 10/27/2005
• Est. Priority Date: 04/22/2004
• Status: Active Grant

• Current Assignee: Fuji Xerox Company Limited (Fujifilm Holdings Corporation)
• Original Assignee: Fuji Xerox Company Limited (Fujifilm Holdings Corporation)
• Inventors: Ashikaga, Hideaki, Kato, Masahiro, Ikegami, Hiroaki, Satake, Masanori, Kimura, Shunichi, Yoshimura, Hiroki, Itonori, Katsuhiko
• Granted Patent: US 7,768,676 B2
• Time in Patent Office: Days
• US Class Current: 382/112
• CPC Class Codes: G06V 10/98 Detection or correction of errors, e.g. by rescanning the pattern or by human intervention; Evaluation of the quality of the acquired patterns

Abstract

An image reading apparatus has a scan quality parameter extraction part, a scan quality evaluation part and a notification part. The scan quality parameter extraction part extracts scan quality parameters representing the quality of a scanned image from image data based on a scan error detection algorithm. The scan quality evaluation part determines whether a scan error occurs, based on the scan error detection algorithm and the scan quality parameters. Also, the notification part notifies a result determined by the scan quality evaluation part to a user.

Claims

1. An image reading apparatus comprising:

a scan quality parameter extraction part that extracts scan quality parameters representing the quality of a scanned image from image data based on a scan error detection algorithm;

a scan quality evaluation part that determines whether a scan error occurs, based on the scan error detection algorithm and the scan quality parameters; and

a notification part that notifies a result determined by the scan quality evaluation part to a user.

2. The image reading apparatus according to claim 1, further comprising:

an automatic document feeder that automatically feeds plural document pages one by one to a document reading location of the image reading apparatus;

a memory that stores page specification information that specifies the page among the plurality of document pages on which it has been determined by the scan quality evaluation part that scan errors have occurred; and

a control part that controls the automatic document feeder to skip the pages up to the page specified by the page specification information.

3. The image reading apparatus according to claim 1, wherein the scan quality evaluation part further calculates scan quality indices that indicate the quality of the scanned image based on the scan quality parameters;

and

the notification part performs notification based on the result determined by the scan quality evaluation part and the scan quality indices.

4. The image reading apparatus according to claim 1, further comprising:

a correctibility determination part that determines whether the scan error is correctable based on an identifier specifying the scan error detection algorithm when the scan error has been detected by the scan quality evaluation part; and

an image processor that corrects image data when the correctibility determination part determines that the scan error is correctable.

5. The image reading apparatus according to claim 1, the scan error detection algorithm comprising:

computing a skew angle of the scanned image, and

comparing the skew angle to a threshold value, wherein the image reading apparatus determines that a scan error occurred when the skew angle is greater than the threshold value.

6. The image reading apparatus according to claim 1, the scan error detection algorithm comprising:

extracting an outline of the document page from the scanned image; and

determining whether the outline is rectangle, wherein the image reading apparatus determines that a scan error occurred when the outline is not rectangle.

7. The image reading apparatus according to claim 1, the scan error detection algorithm comprising:

partitioning the scanned image into plural component images;

computing respective estimated number of characters for the plurality of component images;

performing a character recognition process on the plurality of component images; and

computing a respective character recognition rate for the plurality of component images based on the estimated number of characters and the number of characters recognized by the character recognition process, wherein the image reading apparatus determines that a scan error occurred when the character recognition rates for existing component images are lower than character recognition rates for other component images or a threshold value.

8. The image reading apparatus according to claim 1, the scan error detection algorithm comprising:

performing a rotation operation on the scanned image by zero, 90, 180 and 270 degrees; and

performing a character recognition process on the respective rotated images acquired by the rotation operation, wherein the image reading apparatus determines that a scan error occurred when the number of characters recognized for the character recognition process was at a maximum for a rotated image exclusive of the 0 degree rotated image.

9. The image reading apparatus according to claim 1, the scan error detection algorithm comprising:

converting scan parameters before scanning so as to perform a scan using a size larger than the document size specified by the user; and

determining the existence of significant pixels in regions outside the document size, wherein the image reading apparatus determines that a scan error occurred when it is determined that significant pixels exist in regions outside the document size.

10. The image reading apparatus according to claim 1, the scan error detection algorithm comprising:

applying a Fourier transform to the image data;

computing mean power density for specified high frequency regions among those where power density was acquired by the Fourier transform; and

comparing the mean power density to a threshold value, wherein the image reading apparatus determines that a scan error occurred when the mean power density was greater than the threshold value.

11. The image reading apparatus according to claim 1, the scan error detection algorithm comprising:

converting scan parameter data prior to scanning so as to perform a scan using plural color values without an instruction inputted from the user;

determining whether the scanned data is a color image or a black and white image and a bi-level image or a multi-level image, and

comparing a result determined by a process of determining to the instruction inputted from the user, wherein, the image reading apparatus determines that a scan error occurred when the result is not accordance with the instruction inputted from the user.

12. The image reading apparatus according to claim 1, the scan error detection algorithm comprising:

partitioning every possible value in gradation values of pixels forming the scanned image into plural regions;

measuring the number of pixels having the gradation values that belong to each region;

computing the interval that has the region in which the number of pixels measured exceeds a threshold value; and

comparing the interval of the gradation values with a threshold value, wherein the image reading apparatus determines that a scan error occurred, when the interval is smaller than the threshold value.

13. The image reading apparatus according to claim 1, the scan error detection algorithm comprising:

partitioning the scanned image into plural component images;

performing area calculation for computing the respective areas for the plurality of component images;

performing character recognition for performing character recognition on the plurality of component images;

performing character calculation for computing the number of recognized characters for respective unit areas of the plurality of image components based on the results of the area calculation and the character recognition;

extracting, based on the result of the character calculation, the component image with the largest number of recognized characters per unit area;

applying a Fourier transform to component images extracted at the extracting process;

computing the mean power density of a predetermined frequency band from among the mean power densities calculated by the Fourier transform process;

comparing the mean power density threshold value with the mean power density, wherein the image reading apparatus determines that a scan error occurred when the mean power density is smaller than the mean power density threshold value.

14. An image reading method comprising:

extracting scan quality parameters representing the quality of a scanned image from image data based on a scan error detection algorithm;

determining whether a scan error occurs, based on the scan error detection algorithm and the scan quality parameters; and

notifying a result determined by a process of determining to a user.

15. The image reading method according to claim 14, further comprising:

automatically feeding plural document pages one by one to a document reading location of the image reading apparatus;

storing page specification information that specifies the page among the plurality of document pages on which it has been determined by the process of determining that scan errors have occurred; and

controlling a process of automatically feeding to skip the pages up to the page specified by the page specification information.

16. The image reading method according to claim 14, further comprising:

computing scan quality indices that indicate the quality of the scanned image based on the scan quality parameters; and

notifying the result to the user based on the result determined by the process of determining and the scan quality indices.

17. The image reading method according to claim 14, further comprising:

determining whether the scan error is correctable based on an identifier specifying the scan error detection algorithm when the scan error has been detected; and

correcting image data when a process of determination determines that the scan error is correctable.

18. A storage medium readable by a computer, the storage medium storing a program of instructions executable by the computer to perform a function for image reading comprising:

extracting scan quality parameters representing the quality of a scanned image from image data based on a scan error detection algorithm;

determining whether a scan error occurs, based on the scan error detection algorithm and the scan quality parameters; and

notifying a result determined by a process of determining to a user.