JPH0975860A - Sorting machine, address recognition device and address recognition method - Google Patents

Abstract

(57) [Abstract] [PROBLEMS] To reliably recognize an address letter printed on a postal matter even when the letters are in contact with each other or jammed. SOLUTION: A photoelectric conversion unit 4a for converting a postal matter into image data, and a destination area detection unit 3 for detecting a destination area from the image data of the postal matter converted by the image data conversion unit.
1, character detection cutout unit 34 that detects and cuts out characters in the address area detected by this address area detection unit 31, and character that is cut out by this character detection cutout unit 34 is recognized and recognized. A recognition unit 39 for recognizing the character cut out by the character cutting unit 34 again as an address based on the size of the character in the address area is provided, and the mail piece is classified according to the address recognized by the recognition unit 39. I try to divide it.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a mail sorting machine which sorts mail by using a postal code or an address written on a mail to be conveyed.

[0002]

2. Description of the Related Art Sorting machines for automatically sorting collected mail items to a destination post office are widely used. In order to recognize the address of the mail in this sorting machine, it is necessary to cut out characters from the area where the address is written. If the address is printed type, the size variation of each character is small and they are written at almost constant intervals. Therefore, the boundary between characters can be estimated from the character size and the character arrangement interval. If the size is known, the following character may be cut out according to the size.

Character segmentation and recognition processing in a conventional sorting machine will be described with reference to the block diagram of FIG. That is, the photoelectric conversion unit 11 such as a scanner acquires multivalued image data of the entire image of the mail. Then, the address area detection unit 13 detects an address area from the entire image. Then, the line detection cutout unit 13 cuts out the address line by line from the address region detected by the address region detection unit 13.

The multivalued image data cut out by the line detection / cutout unit 13 is quantized by the quantizing unit 14 into binary image data. Then, the character detection / cutout unit 15 detects the character and cuts out the character.

Thereafter, based on the size of the character detected by the character detection / cutout unit 15, the postal code detection / cutout unit 16 is provided.
The postal code is detected and the postal code is cut out. Then, the postal code recognizing unit 17 recognizes the postal code.

Next, the character recognition unit 18 recognizes the character detected by the character detection / cutout unit 15. Further, the town name / large name recognition unit 19 recognizes the town name / large name.

In addition, the block detection / cutout section 20 detects the block and cuts out the block. Then, the block recognizing unit 21 recognizes the block cut out by the block detecting and cutting unit 20. Furthermore, based on the block recognized by the block recognition unit 21, an answer editing / data collecting unit 22
The data is collected at.

[0008]

The character detection / cutout unit 15 cuts out characters by using projection and labeling. Therefore, as shown in FIG. 6, when the address written on the mail is print type and there is no gap between the characters, if the characters are in contact with each other, the characters cut out by the character detection / cutout unit 15 will be displayed. There is an error in the size of.

For example, in the case of the "Kawa" character of "Kanagawa prefecture" in FIG. 6 (a), if the "Kawa" character has a space, the middle character is separated as shown in a. Therefore, there is a problem that the character cannot be recognized.

Further, when characters are in contact as in "Kosugi-machi" in FIG. 6B, one character is detected in "Kosugi-machi" as indicated by x, so that the block is recognized. There is a problem that the section 21 cannot recognize "Kosugi Town".

In view of the above points, the present invention provides a sorting machine, an address recognizing device, and an address recognizing device capable of surely recognizing the characters even if the address characters printed on the mail are in contact with each other or jammed. To provide an address recognition method.

[0012]

A sorting machine according to a first aspect of the present invention is an image data converting means for converting mail into image data, and a mail converted by the image data converting means. An address area detecting means for detecting an address area from image data, a character cutting means for detecting and cutting out a character in the address area detected by the address area detecting means, and a character cut out by the character cutting means are recognized. Then, the recognition means for recognizing the character cut out by the character cutting means again as the address based on the size of the recognized character in the recognized address area,
The present invention is characterized by including sorting control means for sorting the mail according to the address recognized by the recognition means.

A sorting machine according to a second aspect of the present invention is an image data converting means for converting a mail item into image data, and the image data.
Addressing area detecting means for detecting an addressing area from the image data of the mail converted by the data converting means, character cutting out means for detecting and cutting out characters in the addressing area detected by the addressing area detecting means, and First recognition means for recognizing the characters cut out by the character cutting means, and image data in the address area detected again by the address area detection means based on the character recognition result recognized by the first recognition means. A second recognizing means for recognizing the character and a second recognizing means for recognizing the character cut out by the character cutout portion again based on the size of the character in the address area recognized by the second recognizing means. It is characterized in that it is provided with a second recognizing means and a sorting control means for sorting the mail according to the address recognized by the second recognizing means.

A sorting machine according to a third aspect of the present invention is an image data converting means for converting postal matter into image data, and this image data.
A destination area detecting means for detecting a destination area from the image data of the mail converted by the data converting means, a character cutting means for detecting and cutting out characters in the destination area detected by the destination area detecting means, Determination means for determining whether the address information printed on the mail is printed in print,
When it is determined by this determination means that the address information printed on the mail is printed in print, the characters cut out by the character cutting means are recognized,
Recognizing means for recognizing the characters cut out by the character cutting means again as an address based on the size of the characters in the recognized address area, and sorting control for sorting the mail according to the address recognized by this recognizing means. Means and means are provided.

According to a fourth aspect of the present invention, in the recognition means of the sorting machine, the recognition means recognizes the character at the head of the character cut out by the character cutting means, and the size of the character at the head of the recognized address area. It is characterized in that the character cut out by the character cutting means is recognized again as an address based on the above.

An address recognition apparatus according to a fifth aspect of the present invention comprises an image data converting means for converting a mail into image data, and image data of the mail converted by the image data converting means. The address area detecting means for detecting the address area, the character cutting means for detecting and cutting out the characters in the address area detected by the address area detecting means, and the character cut out by the character cutting means are recognized and recognized. It is characterized by further comprising a recognition means for recognizing the character cut out by the character cutting means again as an address based on the size of the recognized character in the address area.

An address recognizing apparatus according to a sixth aspect of the present invention comprises an image data converting means for converting a mail item into image data, and image data of the mail item converted by the image data converting means. Address area detecting means for detecting the address area, character cutting means for detecting and cutting out characters in the address area detected by the address area detecting means, and first recognition for recognizing the characters cut out by the character cutting means Means and second recognizing means for recognizing the character of the image data in the address area detected by the address area detecting means again based on the character recognition result recognized by the first recognizing means, It is characterized by further comprising: second recognizing means for recognizing the character cut out by the character cutting section again as an address based on the size of the character in the address area recognized by the second recognizing means.

An address recognizing apparatus according to a seventh aspect of the present invention comprises an image data converting means for converting a postal matter into image data and an image data of the postal matter converted by the image data converting means. The address area detecting means for detecting the address area, the character cutting means for detecting and cutting out the characters in the address area detected by the address area detecting means, and the address information printed on the mail are printed in print. If it is determined that the address information printed on the mail is printed in print type by the determination means for determining whether it is a mail, it is cut out by the character cutout means. It is characterized by further comprising a recognition means for recognizing the character and recognizing the character cut out by the character cutting means as the address again based on the size of the recognized character in the recognized address area.

In the address recognizing device according to the present invention, the recognizing means recognizes the character at the head of the character cut out by the character cutting means, and determines the size of the character at the head of the recognized address area. Based on this, the character cut out by the character cutting means is recognized again as an address.

According to a ninth aspect of the present invention, the address recognition method is an image data conversion process for converting a postal matter into image data, and the image data of the postal matter converted in this image data conversion process. The address area detecting step of detecting the address area, the character cutting step of detecting and cutting out the characters of the address area detected in the address area detecting step, and the character cut out in the character cutting step are recognized and recognized. And a recognition step of recognizing the character cut out in the character cutting step as the address again based on the size of the recognized character in the address area.

According to a tenth aspect of the address recognition method of the invention, an image data converting step of converting a mail into image data and an image data of the mail converted in the image data converting step are used. Addressing area detecting step of detecting an addressing area, character cutting step of detecting and cutting out characters in the addressing area detected in this addressing area detecting step, and first recognition of recognizing the characters cut out in this character cutting step And a second recognition step of recognizing the characters of the image data in the address area detected in the address area detection step again based on the character recognition result recognized in the first recognition step. A second recognizing step for recognizing the character cut out by the character cutting section again as an address based on the size of the character of the address area recognized in the second recognizing step.

According to the address recognition method of the eleventh aspect, an image data conversion step of converting a mail item into image data and an image data of the mail item converted in this image data conversion step are performed. The address area detecting step of detecting the address area, the character cutting step of detecting and cutting out the characters in the address area detected in the address area detecting step, and the address information printed on the above mail is printed in print. If it is determined that the address information printed on the mail is printed in print in the determination step for determining whether it is a mail, it is cut out in the character cutting step. It is characterized by further comprising a recognition step of recognizing a character and recognizing the character cut out in the character cutting step as an address again based on the size of the recognized character in the recognized address area.

The present invention is obtained by converting the image data of the mail matter by the image data conversion means (scanner section). Then, the characters in the address area detected by the address area detecting means (address area detecting section) are detected as characters from the image data by the character detecting and cutting means (character detecting and cutting section) and cut out.

Then, the character cut out by the character detecting / cutting-out means is recognized by the recognizing means (character recognizing section), and the character is again recognized based on the size of the character recognized by the recognizing means (character recognizing section). After the character in the address area is cut out by the cut-out means (character detection cut-out part), the cut-out character is recognized as the address by this recognition means. Then, the postal items are sorted according to the address.

Further, the first recognizing means and the second recognizing means are provided as the recognizing means, and the character cut out by the character slicing means is recognized by the first recognizing means to determine the size of the character. To detect. Then, the characters of the image data in the address area detected by the address area detecting means are recognized according to the recognition result by the first recognizing means.

[0026]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a first embodiment of the present invention will be described with reference to the drawings. FIG. 1 shows an automatic mail sorting machine as an address reading sorting machine. In FIG. 1, 1
Is a supply unit in which postal items P such as postcards and sealed letters are set in a lump and in a standing position. The postal matter P set in the supply unit 1 is sequentially taken out one by one from the frontmost postal matter P by the take-out unit 2 and conveyed to the postal matter P by the reading unit 4 arranged in the middle of the conveying path 3. The zip code and address information (address information) that have been sent are read.

The reading unit 4 recognizes a zip code on the postal matter P, a photoelectric conversion unit 4a for photoelectrically converting address information, and a zip code by recognizing a character pattern according to the output of the photoelectric conversion unit 4a. It is a well-known device including a discrimination unit 4b having an address reading device for discriminating address information. Then, the determination unit 4b outputs a classification destination code according to the determined postal code.

Reference numeral 5x designates a postal matter removing unit for eliminating the postal matter P when foreign matter is present in the postal matter P, and reference numeral 5y designates a conveying posture defect accumulating unit for eliminating the postal matter P having a bad conveying posture. .

Next, the postal matter P that has passed through the reading section 4 is conveyed toward the stepped path section 10 through the conveyance path 3. The step pass section 10 is provided with a gate for each of a plurality of steps (steps A to G). The gate mechanism of the step pass section 10 is provided upstream of the gate mechanism. The arrangement of the sensor for detecting the passage of the is similar to the arrangement of the gate mechanism and the sensor in the dividing section 11.

Then, the postal matter P conveyed through the conveying path 3 is branched into any one of a plurality of stages (A stage to G stage) according to the destination code and is conveyed to the sorting section 11. It has become so. Reference numeral 7 is an operator panel provided as an operation panel on the upper portion of the supply unit 1.

The sorting section 11 is provided with a plurality of sorting conveyance paths 5a to 5g in the vertical direction, and the sorting conveyance paths 5a to 5g are selectively sorted according to the sorting destination code. A plurality of sorting boxes 6 ... A plurality of stages (A stages, -G) are provided below the sorting transport paths 5a, 5g along the lower surface side thereof.
The postal matter P, which are arranged in a plurality of stages and are sent to the sorting / conveying paths 5a to 5g, are sorted and collected in a predetermined sorting box 6 based on the reading result (sorting destination code) of the reading unit 4 described above. It is supposed to be done.

A conveyance detector group for detecting the conveyance of the postal matter P on the conveyance path is provided on the conveyance paths 3, 5a, .... Further, near the operation panel 7, a reading position adjuster (not shown) for adjusting the position to be read out of the information written on the mail piece is provided. This reading position adjuster is a card (not shown) on which actual mail is placed.
The reading position is specified by matching with the sol.

Address information and the like are written on the surface of the mail P. The address information is composed of a plurality of lines of character information. As described above, in the postal matter P, the address (address) such as “zip code”, “prefecture name”, “ward, city name”, “town name”, “address” to be read as the address information A (Roppongi 1-1 -1) and company name (○○ corporation) are described.

Next, a control block diagram of the sorting machine will be described with reference to FIG. In FIG. 2, reference numeral 21 is a central control circuit that controls the sorting machine as a whole. The central control circuit 21 is composed of a microprocessor and its peripheral circuits.

The central control circuit 21 controls the above-mentioned feeding unit 1 and the feeding unit control circuit 22 for controlling the conveyance of the postal matter P placed on the feeding unit 1 to the take-out unit 2. The take-out section control circuit 23 that controls the take-out section 2 to take out the postal matter P one by one, the operation display panel control circuit 24 that controls the operation panel 7, and the image of the postal matter displayed on the operation panel 7. Reading position adjustment control circuit 25 for outputting coordinate information designated by the cursor
Is connected.

Further, the central control circuit 21 detects whether or not the foreign matter is present in the postal matter P taken out from the taking-out section 2 and whether or not the posture of the postal matter P being conveyed is good. If there is mail containing foreign matter, the mail P is removed at the branch gate (1), and if the mail is in a poor conveying position, it is discharged at the branch gate (2). Foreign matter / conveyance abnormality detection / elimination circuit 26 for eliminating mail P, sorting module Mn
Control circuit 27, dividing section module stage path section control circuit 28
Is connected.

The sorting module module stage path control circuit 27 receives the mail items conveyed to the stage path unit 10 from the gates A stages to G stages.
The branch gates of the stages are controlled to be opened / closed, and the distribution gates are controlled according to the division destination code.

The sorting module Mn control circuit 28 controls the opening and closing of the gate mechanism for collecting the mails conveyed to each stage in the sorting box. Next, a detailed configuration of the determination unit 4b will be described with reference to FIG. The photoelectric conversion unit 4a acquires multivalued image data of the entire image of the mail. The multi-valued image data of the postal matter P acquired by the photoelectric conversion unit 4a is output to the address area detection unit 31. The address area detection unit detects the address area from the entire image of the mail.

The address area detected by the address area detecting unit 31 is output to the line detection / cutout unit 32. The line detection / cutout unit 32 cuts out the address for each line from the address area detected by the address area detection unit 31.

The multivalued image data cut out by the line detection / cutout unit 32 is output to the quantization unit 33 and quantized into binary image data. Then, the binary image data output from the quantization unit 33 is output to the character detection / cutout unit 34. The character detection / cutout unit 34 detects a character and cuts out the character.

The characters cut out by the character detection / cutout unit 34 are sent to the character recognition unit 35. Character recognition unit 3
A character 5 recognizes a character by comparing the character stored in the dictionary stored in the second dictionary 36 with the character cut out by the character detection cutout unit 34. The character recognizing unit 35 uses the beginning portion of the address, that is, the “prefecture” name
Since only the "city" name is recognized, the capacity of the second dictionary 36 is smaller than that of the first capacity 44 described later.

The characters recognized by the character recognition unit 35 are
It is output to the character detection / cutout unit 34 again. The character detection / cutout unit 34 detects a character from the binary image data output from the quantization unit 33 again based on the recognition result of the character recognition unit 35, and cuts out the character.

Then, based on the size of the character detected by the character detection / cutout unit 34, the postal code detection / cutout unit 37 is provided.
The postal code is detected and the postal code is cut out. Then, the cut out postal code is recognized by the postal code recognition unit 38.

Next, the character recognition unit 39 recognizes the character detected by the character detection / cutout unit 34. Further, the town name / large name recognition unit 40 recognizes the town name / large name.

In addition, the block detection / cutout unit 41 detects the block and cuts out the block. Then, the block recognizing unit 42 recognizes the block cut out by the block detecting and cutting unit 41.

Further, based on the block recognized by the block recognizing unit 42, the answer editing and data collecting unit 43 collects data. The postal code recognition unit 38, the character recognition unit 39, the town name / large name recognition unit 40, and the block recognition unit 4
2 includes a first dictionary 44 that stores a character pattern that serves as a reference for recognizing characters. By the way, the capacity of the second dictionary 36 is smaller than that of the first dictionary 44.

Next, the operation of the first embodiment of the present invention configured as described above will be described. The postal items P such as postcards and envelopes are set in a batch and standing state, and are sequentially taken out one by one from the frontmost postal item P by the take-out unit 2 and conveyed through the conveyance path 3. .

Then, it is sent to the reading section 4 located on the way of the transport path. In the photoelectric conversion unit 4a of the reading unit 4, the postal code and address information on the postal matter P are photoelectrically converted as multi-valued image data.

The multi-valued image data of the postal matter P is sent to the address area detection unit 31 and the address area is detected from the entire image of the postal matter. Then, the line detection cutout unit 32 cuts out the address for each line from the address area detected by the address area detection unit 31.

The multivalued image data cut out by the line detection / cutout unit 32 is output to the quantization unit 33 and quantized into binary image data. Then, the binary image data output from the quantization unit 33 is output to the character detection / cutout unit 34, and the character is detected and the character is cut out.

The operation of the character detection / cutout unit 34 will be described in detail with reference to the flowchart of FIG. That is, the character detection / cutout unit 34 performs projection and labeling processing on the binary image data quantized by the quantization unit 33 (step S11). Then, the space between characters, the pitch, and the like are calculated from the results of the projection and labeling processing (step S12).

Then, a candidate point for cutting out a character is calculated from the space, pitch, etc. between the characters calculated in step S12 (step S13). That is, as shown in FIG. 6, "Kanagawa Prefecture""3-441 Kosugicho, Nakahara-ku, Kawasaki"
If there is such an address, the four candidate inspection points a to d and A to D are calculated by calculating the character inspection candidate points in step S13.

Then, the process of selecting the cut-out candidate points from the cut-out candidate points a to d and A to D, that is, the cut-out candidate points are selected from the four candidate points as the primary check-out candidate. (Step S14).

Next, as a reading mode, a print type mode is used.
Mode is determined (step S1)
5). In the determination of step S15, "YES"
If it is determined that the characters are recognized as the primary candidates by the cut-out candidate points selected in step S14, the character recognition unit 35 performs the processing (step S16). The character recognition unit 35 searches for "prefecture", "city", "ward", etc. as delimiter characters (also called delimiters) at the beginning of the address. In other words, by searching for a place name such as "Kanagawa Prefecture" or "Kawasaki City", "prefecture" as a delimiter,
Searching for "city" and "ward" (step S17).

Then, the character recognition unit 35 performs a process of returning the character size information and the number-of-characters information of the detected delimiter characters "prefecture", "city" and "ku" to the character detection / cutout unit 34 (step S18).

The character detection / cutout unit 34 detects the character from the binary image data output from the quantization unit 33 again based on the recognition result of the character recognition unit 35, and cuts out the character. There is.

Then, based on the size of the character detected by the character detection / cutout unit 34, the postal code detection / cutout unit 37 is provided.
The postal code is detected and the postal code is cut out. Then, the cut out postal code is recognized by the postal code recognition unit 38.

Next, the character recognition unit 39 recognizes the character detected by the character detection / cutout unit 34. Further, the town name / large name recognition unit 40 recognizes the town name / large name.

In addition, the block detection / cutout unit 41 detects the block and cuts out the block. Then, the block recognizing unit 42 recognizes the block cut out by the block detecting and cutting unit 41.

Further, based on the block recognized by the block recognition unit 42, the answer editing and data collection unit 43 collects data. In addition, by searching for a place name such as "Kanagawa Prefecture" or "Kawasaki City" in step S17, if "prefecture", "city", or "ward" as a delimiter is not found, a new By finding such a candidate, the first character can be detected in "Kanagawa Prefecture" and "Kawasaki City" in step S17.

As described above, in the first embodiment, the character detection / cutout unit 34 recognizes the cut-out address by the character recognition unit 35.
The character size obtained as a result is sent back to the character detection cutout unit 34. Therefore, when the character detection cutout unit 34 performs the detection cutout again, the address written in the mail P is The cutout can be made accurate.

Therefore, the address can be correctly recognized even when the characters written on the mail contact. Further, since the character recognition unit 35 only recognizes the character at the beginning of the address cut out by the character detection / cutout unit 34, the capacity of the second dictionary to be used can be reduced.

Next, a second embodiment of the present invention will be described. In the above-described first embodiment, the character detected by the character detection / cutout unit 34 is recognized by the character recognition unit 35 provided separately from the character recognition unit 39, and this recognition unit 35 is recognized.
The characters are again cut out by the character detection / cutout unit 34 based on the recognition result in the above, but in the second embodiment, as shown in FIG. 5, the character recognition unit 39 and the town name / large name recognition unit 40
By feeding back the recognition result of No. 1 to the character detection / cutout unit 34, the same processing as that of the first embodiment is performed in step S17 of the flowchart in FIG.

As described above, in the second embodiment, it is not necessary to provide the special character recognition unit 35 as in the first embodiment. In the above-described two embodiments, the delimiter characters “prefecture” and “city” are recognized and the size of the delimiter character is set to the size of each character. However, the delimiter character is not limited to other characters. Anything that you can recognize is fine.

[0065]

According to the invention described in claims 1 to 11, when a character is detected by dividing an address described in a mail, the character is recognized once and the size of the character is recognized. Therefore, since the characters are separated again, the sorting machine, address recognition device and address recognition that can reliably recognize the characters even if the address characters printed on the mail contact or are jammed To provide a method.

[Brief description of drawings]

FIG. 1 is a diagram showing an automatic mail sorting machine according to a first embodiment of the present invention.

FIG. 2 is a control block diagram of the sorting machine.

FIG. 3 is a diagram showing a detailed configuration of a discriminating unit.

FIG. 4 is a flowchart for explaining the operation of the first embodiment.

FIG. 5 is a block diagram showing a configuration of a discriminating unit according to a second embodiment of the present invention.

FIG. 6 is a diagram showing an address for explaining an operation.

FIG. 7 is a diagram showing a configuration of a conventional determination unit.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Supply part, 2 ... Take-out part, 3 ... Conveyance path, 4 ... Reading part, 4
a ... Photoelectric conversion unit, 4b ... Discrimination unit.

Claims (11)

[Claims] 1. Image data for converting a postal matter into image data
Data conversion means and the image data of the mail converted by the image data conversion means.
Address area detecting means for detecting the address area from the data, character cutting means for detecting and cutting out the characters in the address area detected by the address area detecting means, and recognizing and recognizing the characters cut out by the character cutting means. Based on the size of the recognized character in the address area, the recognition means for recognizing the character cut out by the character cutting means again as an address, and the mail piece are classified according to the address recognized by this recognition means. A sorting machine comprising: sorting control means. 2. Image data for converting postal matter into image data
Data conversion means and the image data of the mail converted by the image data conversion means.
Address area detecting means for detecting the address area from the data, character cutting means for detecting and cutting out the characters in the address area detected by the address area detecting means, and first recognizing the character cut out by the character cutting means.
And second recognizing means for recognizing the characters of the image data in the address area detected by the address area detecting means again based on the character recognition result recognized by the first recognizing means. The second recognizing means and the second recognizing means for recognizing again the character cut out by the character cutting section as the address based on the size of the character in the address area recognized by the second recognizing means. A sorting machine comprising: sorting control means for sorting mail according to the recognized address. 3. Image data for converting a postal matter into image data
Data conversion means and the image data of the mail converted by the image data conversion means.
Address area detecting means for detecting the address area from the address area, character cutting means for detecting and cutting out the characters in the address area detected by the address area detecting means, and the address information printed on the above-mentioned mail in printed form. If it is determined by the determining means that the address information printed on the mail is printed in print, the determining means determines whether the address information is printed. Recognized characters,
A recognition means for recognizing again the characters cut out by the character cutting means based on the size of the characters in the recognized address area, and a sorting control for sorting the mail according to the addresses recognized by this recognition means. And a sorting machine. 4. The recognizing means recognizes the character at the head of the character cut out by the character cutting means, and again based on the size of the character at the head of the recognized address area, the character cutting means again. 4. The sorting machine according to claim 1, wherein the character cut out in step 4 is recognized as an address. 5. Image data for converting postal matter into image data
Data conversion means and the image data of the mail converted by the image data conversion means.
Address area detecting means for detecting the address area from the data, character cutting means for detecting and cutting out the characters in the address area detected by the address area detecting means, and recognizing and recognizing the characters cut out by the character cutting means. An address recognition device comprising: a recognition unit that recognizes the character cut out by the character cutting unit again as an address based on the size of the recognized character in the address area. 6. An image data for converting a postal matter into image data.
Data conversion means and the image data of the mail converted by the image data conversion means.
Address area detecting means for detecting an address area from the data, character cutting means for detecting and cutting out characters in the address area detected by the address area detecting means, and first recognizing the character cut out by the character cutting means
And second recognizing means for recognizing the characters of the image data in the address area detected by the address area detecting means again based on the character recognition result recognized by the first recognizing means. A second recognizing means for recognizing the character cut out by the character cutting section again as an address based on the size of the character in the address area recognized by the second recognizing means. Address recognition device. 7. Image data for converting postal matter into image data
Data conversion means and the image data of the mail converted by the image data conversion means.
Address area detecting means for detecting the address area from the address area, character cutting means for detecting and cutting out the characters in the address area detected by the address area detecting means, and the address information printed on the above-mentioned mail in printed form. If it is determined by the determining means that the address information printed on the postal matter is printed in print, it is cut out by the character cutting means. Recognized characters,
An address recognition apparatus comprising: a recognition unit that recognizes the character cut out by the character cutting unit again as an address based on the size of the recognized character in the recognized address area. 8. The recognizing means recognizes the character at the head of the character cut out by the character cutting means, and again based on the size of the character at the head of the recognized address area, the character cutting means again. 8. The address recognition device according to claim 5, wherein the character cut out in step 4 is recognized as an address. 9. An image data for converting a mail matter into image data.
Data conversion process and the image data of the mail converted in this image data conversion process.
Address area detection step of detecting the address area from the data, character cutting step of detecting and cutting out the characters of the address area detected in this address area detection step, and recognizing and recognizing the character cut out in this character cutting step. And a recognition step of recognizing again the character cut out in the character cutting step as an address based on the size of the recognized character in the address area. 10. An image data conversion process for converting a mail matter into image data, and image data of the mail matter converted in this image data conversion process.
The address area detecting step of detecting the address area from the data, the character cutting step of detecting and cutting out the characters of the address area detected in the address area detecting step, and the first step of recognizing the character cut out in the character cutting step.
And a second recognition step of recognizing the characters of the image data in the address area detected again in the address area detection step based on the character recognition result recognized in the first recognition step. And a second recognizing step for recognizing the character cut out by the character cutting section again as an address based on the size of the character in the address area recognized in the second recognizing step. Address recognition method. 11. An image data conversion process for converting a mail into image data, and image data of the mail converted in this image data conversion process.
Address area detection step of detecting the address area from the data, character cutting step of detecting and cutting out the characters of the address area detected in this address area detection step, and the address information printed on the above mail is printed in print. If it is determined in this determination step that the address information printed on the postal matter is printed in print, it is cut out in the character cutting step. Recognized characters,
An address recognition method comprising: a recognition step of recognizing a character cut out in the character cutting step again as an address based on the size of the recognized character in the recognized address area.