JPH08243507A - Mail processing device and image display method - Google Patents

Abstract

(57) [Summary] [Purpose] In a video coding system in which an operator reads a postal image and inputs a postal code, etc., the image is displayed so that necessary information can be read more easily. When a postal image 201 is displayed on a display device 105, an input target character cutout image is displayed in a partial image display area 202 based on information obtained by a recognition unit, and the character recognition result is regarded as a character image. Echo back area 2 so that correspondence can be taken
03 is displayed. When displaying a plurality of postal images, the postal images 201 are displayed in an overlapping manner, and the partial image display area 202 and the echo back area 203 are displayed so as not to overlap. [Effect] The information to be input can be found early in the mail image, and the input efficiency of the operator is improved.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of displaying a recognition result in an information medium such as a document which is divided into a plurality of partial areas and is substantially stylized, and corresponds to an image of the information medium read by a scanner or the like. The present invention relates to a video coding device that inputs a code. Particularly, it is suitable for a mail processing device that scans a destination part such as a mail with a scanner or the like, an operator inputs a code corresponding to the address based on the mail image, and processes the mail based on the input code. Is.

[0002]

2. Description of the Related Art A video coding apparatus is used for the purpose of recognizing an information medium such as a document and improving the efficiency of processing corresponding to the recognition result. The video coding device makes it possible to improve the efficiency of mail sorting work in the post office. In this case, the video coding device reads the destination part of the mail as image data, displays the read image data on a CRT or the like, and the operator reads and inputs the destination information such as the zip code from the displayed contents and inputs it. The mail is processed based on the information obtained. Specific contents of processing include sorting the mail items mechanically based on the input information, converting the input information into mechanically readable information such as barcodes, and printing it on the mail items. .

At present, a method of mechanically recognizing the destination information from the read image data and processing the mail based on the result is also put into practical use. This method does not require manual labor for input and also increases the processing speed. However, due to problems such as the quality of the characters in the destination portion and the quality of the read image data, some postal items cannot mechanically recognize the destination information. A method combining these methods has also been put into practical use. That is, first, the destination information is mechanically recognized, and if the destination information cannot be recognized, a person looks at the corresponding postal image and inputs the destination information. At this time, a method is considered in which a person does not input all the destination information described in the postal image, but only the characters that cannot be mechanically recognized. For example,
Assuming that the destination information is the postal code “213-45”, and when only the second character “1” is unrecognizable as a result of machine recognition, a person inputs only “1”.
Hereinafter, this method is referred to as a "partial input method".

When a person such as the partial input system reads address information from a postal image displayed on a CRT or the like and inputs it from a keyboard or the like, how to display the postal image becomes a problem. By displaying the destination information in a format that is easier to read, input efficiency is improved. JP-A-59-145
Japanese Patent Publication No. 084 discloses a method in which a destination area is cut out and only that portion is displayed with a certain size, so that even if the size of the destination character on the postal matter is small, it is displayed on the screen with certainty. ing. Japanese Unexamined Patent Publication No. 57-146377 discloses a method in which two states, that is, a state in which the entire image is displayed and a state in which only the destination area is displayed, are prepared and can be switched. In addition, JP-A-6-2
Japanese Patent No. 06051 discloses a method in which an area in which a postal code is described is cut out and displayed so as to be superimposed on the upper left portion of the entire image.

[0005]

Here, in the case of the "partial input method", it is necessary to read from the displayed image which character is the input target character (character that could not be mechanically recognized). . It is effective to increase the input efficiency by clearly indicating to the operator how much character is currently input and which character is input next.

[0006] However, in any of the above-mentioned conventional techniques, there is no awareness of the problem with respect to clearly indicating the character to be input next.

Further, in order to pre-read the input target character in the image such as the address of the mail, a plurality of images are
There is a method to display it on the screen. However, due to the limitation of the screen size, usually only about two images can be displayed.
With this number of sheets, it is not possible to provide the operator with sufficient information for prefetching. Considering that necessary information is prefetched from the image, it is desired that the number of images displayed on one screen is large to some extent.

[0008]

According to the present invention, each input (recognition) target character area information output from the character recognition section is used. Here, the target character area information refers to coordinate information such as the position of the recognition target character in the entire image, the character type and size of the recognition target character, and the characteristics thereof. Then, the character images to be input are copied from the read image and are collectively displayed at a predetermined position on the screen. An input echo back area is displayed near this input target character image. The character recognition result output from the character recognition unit is displayed in the input echo back area. At this time, the unrecognized characters are displayed in an explicit form (for example, the unrecognized characters are displayed in a predetermined symbol or in a character type different from the recognized character type of the character that seems to be closest to the result) and input. Each character in the target character image and each character in the input echo back area are displayed clearly associated with each other. The characters input from the input device are displayed by being assigned to the "unrecognized characters" portion in the input echo back. Further, when displaying a plurality of images on the screen, the images can be displayed in an overlapping manner, and the input target character image (partial image in each image) and the input echoback area corresponding to each image can be displayed without overlapping. To arrange.

[0009]

When the character region is accurately cut out from the read image, the operator does not look for the input target character in the entire image, while looking at the input target character image displayed at a predetermined position, Input processing can be performed. Further, even in the case of the input method of inputting only the character that the recognition device cannot recognize, it is possible to instantly determine which character in the image is the current input target. In addition, since a plurality of postal images are displayed in the range where the input character image and the input echoback area do not overlap, it is possible to input while pre-reading the images after the current input target, and a faster input operation can be performed.

[0010]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT An embodiment of the present invention will be described below with reference to FIGS.

FIG. 1 shows the overall configuration of a video coding apparatus according to this embodiment. The image reading unit 101 reads the address description surface of the postal matter 100, and the recognition unit 102 recognizes the destination information from the image data. When the destination information cannot be recognized, the corresponding postal image is video coding desk (hereinafter referred to as VCD) 1 via the communication path 103.
Sent to 04. The sent mail image is displayed on the display device 105, the VCD operator reads the destination information, and inputs the destination information by the input device 106. Display device 10
5 and the input device 106 are controlled by the control unit 107. The devices in the VCD 104 can exchange information with each other through the bus 108. The mail processing unit 109 is
The destination information generated by the recognizer 102 or VCD 104 is used to process the corresponding mail piece 100.
The processing content of the mail processing unit 109 is 100
Or the like, or the destination information is converted into mechanically readable information (bar code or the like) and printed on the postal matter 100. Image reading unit 101 and recognition unit 1
Each device such as 02 is controlled by the control unit 110 and can exchange information with each other through the bus 111. The destination information may be a postal code, a destination address, a code corresponding to the destination address, a number (chome address) included in the destination address, or the like. In the following examples, the destination information to be coded is the postal code.

FIG. 2 is an example of a postal image display on the display device 105 of the VCD 104. In this display example, four postal images (201a to 201d) are displayed horizontally.
The partial image areas (202a-
202d). An image obtained by cutting out the postal code portion from the corresponding postal image (201a to 201d) is displayed in this area. Below that, the echo back area (203
a to 203d) are provided. In the echo back area, the result of character recognition of the postal code part from the corresponding postal image is displayed in characters. For the unrecognized characters,
In the example of 203a, the third character and the fifth character cannot be recognized. For the second postal image 201b, the postal code part cannot be cut out and recognized. , Area 202b and area 203
b is blank. In this screen display state, the postal image 201a displayed at the top is a coding target. One of the features of the present invention is that the partial image area 202a is
To 202d, the echo back areas 203a to 203d are displayed in correspondence with the character images displayed on the screens to 202d.
This makes it easier to check the unrecognized characters. Another feature is that the partial image areas 202a to 202d and the echo back areas 203a to 203d are displayed so as not to overlap even if the postal images 201a to 201d are displayed in an overlapping manner. This makes it possible to pre-read necessary information (image of postal code and recognition result).

FIG. 3 is a processing flow of the control unit 110. Each process will be described.

In process 301, the image reading unit 101 reads the image of the surface on which the address of the postal matter is described.

In process 302, the recognition unit 102 recognizes the destination information (postal code) based on the image read in process 301 to generate a code.

A process 303 determines whether or not the recognition is successful in the process 302. "Success of recognition" here means
The address information (zip code) can be completely recognized and can be generated as a code. If the recognition is successful, process 3
If No. 05 is performed and the recognition is not successful, the process 30 is performed.
Do 4.

In step 304, mail code that cannot be recognized by the recognition unit 102 is video coded to generate a code of address information. After this process, process 305 is performed. Details of the process 304 will be described later with reference to FIG.

Finally, in process 305, the mail processing unit 1
Using 09, the postal matter is processed based on the code of the destination information generated in the process 302 or the process 304. As the contents of processing the mail, it is possible to sort the mail according to the destination, convert the code (address information) into mechanically readable information (bar code, etc.), and print it on the mail.

FIG. 4 is a processing flow of the processing 304 (video coding). This processing flow is VCD1
It is controlled by the control unit 107 of 04. Each process will be described.

In process 401, the image read in process 301 is displayed on the display device 105. Details of this processing 401 will be described later with reference to FIG.

Then, in process 402, the VCD operator views the postal image displayed in process 401 and inputs the corresponding destination information from the input device 106.

FIG. 5 is a processing flow of processing 401 (image display). FIG. 6 is an example of information used in the process (process 401) of FIG. The information used in the process 401 includes the recognition information 601 generated in the process 302 and the postal image 602 generated in the process 301. Recognition information 6
The contents of 01 are "number of characters 603", "character area 6"
04 ”,“ upright direction 605 ”,“ recognition result 606 ”4
There is one. The number of characters 603 is the number of characters of the destination information.
In this example, the destination information is a five-digit postal code, so the number of characters = 5. The character area 604 is a display area of each character of the destination information on the postal image 602. In this example, there are a1 to a5 as the character area 604, and the area of each character (604
a to 604e). The erect direction 605 indicates the erect direction of the destination information. In this example, it becomes erect when rotated 90 degrees to the left. The recognition result 606 is code information that was partially recognized in the process 302. In this example,
The character string "13 # 7 #" is regarded as the "recognition result" because the third character and the fifth character cannot be recognized when "13579" is recognized. Taking the information in FIG. 6 as an example,
Each process of FIG. 5 will be described.

In process 501, the postal image 602 is displayed on the display device 105. FIG. 7 shows a screen display example at this point.

In step 502, it is determined whether or not there is character area information. If there is character area information, the process 503 is performed, and if not, the process 401 ends. In this example,
Since the character area information 604 exists, processing 503 is performed.

In process 503, the postal code part is cut out from the postal image 602 and displayed by using the character area information determined in process 502. FIG. 8 shows a screen display example after this processing. In this screen display example, an image obtained by cutting out the postal code portion is displayed in the partial image area 202. Figure 9
Is a processing flow of the processing 503. Each process will be described.

In process 901, the value of the variable i is set to 1.

In process 902, the area of the i-th character of the postal code is cut out from the postal image 602. In FIG. 10, processing 90
An outline of processing after 2 is shown. FIG. 10 shows an example when i = 1. An area 604a for the first character of the character area 604 is cut out, and this is set as an image x (1001).

In the process 903, the image x (1001) is rotated in the upright direction 605, and the result is rotated to the image y (100).
2). In the example of FIG. 10, the upright direction 605 is “left 9
Since it is "0 degree", the image y (10
02) is generated.

In process 904, the image y (1002) is
Enlargement / reduction is performed so as to have a constant size, and the result is image z (1003). In the example of FIG. 10, the image y (1
002) is enlarged so as to have a size obtained by dividing the partial image area 202 into five equal parts, and an image z (1003) is obtained.

In process 905, the image z (1003) is
It is displayed at the i-th character in the partial image area 202. In the example of FIG. 10, the image z (1003) is displayed on the leftmost side which is the area of the first character. In the case of the second character, it is displayed to the right of it.

In process 906, the variable i is incremented by 1.

Finally, in process 907, it is determined whether the variable i is less than or equal to the number of characters 603 in the postal code. If the determination result is true, the process returns to step 902. If the determination result is false, step 503 is terminated and step 504 is performed.

In step 504, the echo back area is displayed. FIG. 11 shows a screen display example after this processing. In this screen display example, the echo back area 203 is displayed below the partial image area 202, and the character string “13 # 7 #” which is the recognition result of the postal code is displayed here. FIG. 12 shows internal information used in the process 504. The internal information 1201 is “the number of unread characters 1202”,
There are four points: "unread position 1203", "input character number 1204", and "input content 1205". Unread characters 1202
Is a variable that represents the number of characters that cannot be recognized by the recognition unit 102, and the unread position 1203 is a variable that represents the position of the unrecognized character (position from the beginning). Since the recognition result 606 of this embodiment is “13 # 7 #”, the number of unread characters = 2, unread position (1) = 3, and unread position (2) = 5. The number of input characters 1204 is a variable indicating the number of characters input by the VCD operator, and the input content 1205 is a variable for storing the input characters on a character-by-character basis. In the initial state, the number of input characters = 0, the input content (1) = NULL, and the input content (2) = NULL.

FIG. 13 is a processing flow of the processing 504. Each process will be described.

In process 1301, the postcode recognition result 606 is set in the variable "echoback character string", and the number of characters 603 of the recognition result 606 is set in the variable "echoback character number". In this embodiment, the echo back character string is 13 #
7 #, the number of echo back characters is 5.

In process 1302, the value of the variable i is set to 1.

In process 1303, it is determined whether the value of the variable i is the number of input characters or more. If the determination result is true, processing 1306 is performed, and if false, processing 1304 is performed.
In this example, since the input operation has not been performed yet and the number of input characters is 0, processing 1306 is performed.

In the process 1304, the "unread position (1203) (i)" character of the echo back character string is changed to the input content (1205) (i).

In process 1305, the value of the variable i is incremented by 1. After this processing, the procedure returns to processing 1303. Process 1303
By processing 1305, when a character is input from the input device 106, the result is reflected in the echo back character string.

In the process 1306, the value of the variable i is set to 1.

In process 1307, the i-th character of the echo back character string is displayed immediately below the character image displayed in the i-character of the partial image area 202. For example, if i = 3,
The character "#" is displayed immediately below the image "5" corresponding to the third character in the partial image area 202.

In the process 1308, the value of the variable i is incremented by 1.

In process 1309, it is determined whether the value of the variable i is equal to or larger than the number of echo back characters. If the determination result is true, the process 504 is terminated, and if false, the process returns to the process 1307.

When the process 504 is completed, the process 401
Ends. At this point, the image display ends, and the process 402 which is an input (coding) process is performed.

FIG. 14 is a processing flow of the processing 402 (coding). Each process will be described.

In processing 1401, the echo back area 20
The character at the unread character position that is the current input target in 3 is highlighted. In the initial state, the number of input characters 1204 =
Since it is 0, the first unreadable character, that is, the zip code 3
The character (unread position 1203 (1) = 3) is highlighted. FIG. 15 shows a screen display example at this point. A cursor 1501 representing the current input target character is displayed at the third character in the echo back area 203. With the display of the cursor 1501, it is possible to instantly read the position of the character that needs to be input at present and the character image corresponding to the position.

In process 504, the same process as process 504 described above is performed.

In process 1402, it is determined whether the "confirmation key" of the input device 106 has been pressed. If it is pressed, it is considered that the input is completed, and the process 402 is ended. If it has not been pressed, processing 1403 is performed.

In process 1403, it is determined whether the "one character cancel key" is pressed and the number of input characters 1204 exceeds 0. If the determination result is true, process 1404 is performed, and if false, process 1405 is performed. The “single-character cancel key” is a key existing in the input device 106.

In process 1404, the number of input characters 1204 is decremented by 1. This will cause the last 1 of the entered string to
The character has been deleted. After this processing, the processing 140
Return to 1.

In process 1405, it is determined whether or not there is a key input from the input device 106 and the number of input characters 1204 is 1202 or less. If the determination result is true, the process 1406 is performed, and if the determination result is false, the process returns to the process 1401.

In process 1406, the contents keyed in from the input device 106 are stored in the input contents 1205, and the value of the number of input characters 1202 is incremented by 1. “5” in processing 1405
When is input, the input content (1) = 5, the number of input characters =
It becomes 1. After this process, the process returns to the process 1401.

FIG. 16 shows an example of a screen display at the time when the processing 1401 and the processing 504 are performed after "5" is input in the processing 1405. The input character "5" is displayed at the position where the cursor 1501 was displayed in the stage of FIG. Since the number of input characters 1204 has increased by 1, the cursor 1501 is displayed at the second unread character in the echo back area 203, that is, at the fifth character of the postal code (unread position 1203 (1) = 5).

FIG. 17 shows a screen transition example when four postal images (201a to 201d) are displayed on the display device 105. Further, FIG. 18 shows a flow of the video coding 304 corresponding to this screen transition. Each process in FIG. 18 will be described together with the screen transition example in FIG.

In process 1801, it is determined whether the value of the variable count is 0 or less. If the judgment result is true, process 4
01a, and if false, process 402 is performed. Variable c
"outt" is a variable indicating the number of postal images 201 displayed on the screen. In the example of FIG. 17, count = 4 in the state of the screen 105a, and count = 3 in the state of the screen 105b.

In process 401a, four postal images 20
1, partial image area 202 and echo back area 203
Is displayed. The content of the processing is similar to that of the above-described processing 401 performed four times. Screen 105a and screen 1 in FIG.
05e is the state immediately after the processing of this processing 401a.

In process 1802, the value of the variable count is set to 4.

In process 402, the same process as the process 402 described above is performed.

In process 1803, the image currently being coded is erased from the display device 105, and the value of the variable count is reduced to 1. In the example of FIG. 17, the screen 105a is the screen 10
5b, screen 105b, screen 105c, screen 105c
Changes to screen 105d. In the case of the screen 105d, all postal images are once erased by the process 1803, but when the process 304a is performed thereafter, the judgment of the process 1801 becomes true, and the screen 105e is obtained through the process 401a.

The screen transition based on the process 304a is shown in FIG.
As shown in the example of No. 7, the display position of each image is fixed, and the display position does not change every time one coding is completed. Furthermore, since the partial image area 202 of the postal code and the echo back area 203 in which the recognition result is initially displayed are displayed without overlapping, prefetching is easy and improvement in input efficiency can be expected.

FIG. 1 based on the processing flow 304a of FIG.
In addition to the screen transition of 7, some screen transitions are possible. 19, 20, and 21 are examples of screen transitions different from those in FIG. In the example of FIG. 19, the screen transitions so that the current input target postal image is always displayed at the lower left of the display device 105. For example, when the coding of the postal image 201a is completed in the state of the screen 105a in FIG. 19, the postal image 201a disappears as shown in the screen 105f, the postal images 201b to 201d are shifted one by one in the lower left, and a new postal image 201e is in the upper left. Is displayed in.

In the examples of FIGS. 20 and 21, the screen transitions so that the current input target postal image is displayed above other postal images. For example, when the coding for the postal image 201a is completed in the state of the screen 105a in FIG. 20, the postal images 201a to 201d are still displayed and the postal image 201b is displayed at the top so as not to overlap with other postal images. Is displayed.

In the embodiments up to FIG. 21, the destination information to be coded is the postal code, but other information may be coded as the destination information. FIG.
In addition to the postal code, a screen display example in the case of coding a town name and a house number (a number indicating a chome, address, number, etc.) as destination information is shown. In the example of FIG. 22, the display device 105
Besides the postal image 602, the partial image area 202, and the echo back area 203, the town name image 21 in the postal image 602 is displayed.
There is a town name partial image area 202x displaying 01 and a house number partial image area 202y displaying a house number image 2102. Below the town name partial image area 202x, a town name echo back area 203x for displaying the result of recognition of the town name image is displayed.
Below the house number partial image area 202y, there is a house number echo back area 203y for displaying the result of recognizing the house number image. In the example of FIG. 22, after the town name expressed as “ΔΔ town” is recognized, it is converted into the numerical code “012” corresponding to “ΔΔ town” and displayed in the town name echo back area 203x. There is. Town name partial image area 2
02x and the house number partial image area 202y can be displayed by the same method (process 503) as the partial image area 202. The town name echo back area 203x and the house number echo back area 203y can be displayed by the same method (process 504) as the echo back area 203.

[0064]

According to the present invention, when displaying a postal image, a character image directly related to coding (a postal code in the embodiment) is cut out and displayed, and a recognition result corresponding to each is displayed in association with each other. To do. Therefore, the operator who looks at the postal image and codes the necessary information can quickly find the position of the character that must be input and the corresponding character image. Further, even when displaying a plurality of postal images such as four, the cut-out character image and the corresponding recognition result can be viewed on one screen without overlapping, so that the information required for input is pre-read. be able to. Further, by displaying not only the character image cutout result and the recognition result but also the entire mail image, the entire image can be viewed as it is even for the mail item for which the character cutout has failed.

[0065]

[Brief description of drawings]

FIG. 1 is an overall configuration diagram of a video coding apparatus in this embodiment.

FIG. 2 is a mail image display example on a VCD display device.

FIG. 3 is a processing flowchart of a system control unit.

FIG. 4 is a processing flowchart of a VCD control unit.

FIG. 5 is a processing flowchart of image display processing.

FIG. 6 is a diagram showing information used in image display processing.

FIG. 7 is a diagram showing a screen in a mail image display state.

FIG. 8 is a diagram showing a screen after the destination character cutout display processing.

FIG. 9 is a processing flowchart of destination character cutout display processing.

FIG. 10 is a conceptual diagram of destination character cutout display processing.

FIG. 11 is a diagram showing a screen after echo back area display processing.

FIG. 12 is a diagram showing information used in echo back area display processing.

FIG. 13 is a processing flowchart of echo back area display processing.

FIG. 14 is a processing flowchart of coding processing.

FIG. 15 is a diagram showing a screen (initial state) in an input target character highlighting state.

FIG. 16 is a diagram showing a screen in which an input target character is highlighted (one character input state).

FIG. 17 is a screen transition diagram (part 1) when four postal images are displayed.

FIG. 18 is a processing flowchart of coding processing when four postal images are displayed.

FIG. 19 is a screen transition diagram (part 2) when four postal images are displayed.

FIG. 20 is a screen transition diagram (part 3) when four postal images are displayed.

FIG. 21 is a screen transition diagram (part 4) when four postal images are displayed.

FIG. 22 is a screen display example for coding a postal code, a part of an address, and a house number.

[Explanation of symbols]

101 ... Image reading unit, 102 ... Recognition unit, 105 ... Display device, 106 ... Input device, 107 ... VCD control unit, 10
9 ... Mail processing unit, 110 ... System control unit

Claims (3)

[Claims] 1. An image reading unit for optically reading a postal image from the address description surface of a postal matter; a recognition unit for recognizing information necessary for processing the postal image read by the image reading unit; Display means for displaying the information that could not be recognized by the recognition means in association with the postal image of the area to which the information belongs, and input means for the operator to input the unrecognized information displayed on the display means, A mail processing apparatus, comprising: processing means for processing the mail using the information input by the input means and the information recognized by the recognition means. 2. The mail processing apparatus according to claim 1, wherein the display unit displays a plurality of the mail images in duplicate, and the unrecognized information corresponding to each of the images and the mail of the area to which the information belongs. A postal image display device characterized in that each set of images is displayed while being shifted. 3. At least one of a character and a symbol,
Optically reading an information recording medium described in a substantially stylized form as an image, recognizing desired information in the image, and displaying a predetermined partial image of the image and a recognition result of the partial image in association with each other. A display method characterized by the above.