JP4261988B2 - Image processing apparatus and method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an image processing apparatus for retrieving image data similar to an image read by an input device such as a scanner from a database and outputting it.
[0002]
[Prior art]
2. Description of the Related Art In recent years, an image processing apparatus has been proposed in which a paper document or distributed material stored in a binder or the like is read by a scanner and original electronic data is searched. Patent Document 1 expands an electronic document in a database into a raster image, compares it with a scanned image, narrows down search results, compares the document with the highest similarity with a predetermined reference value, and exceeds the reference value. Then, the document is output to the display unit, and then printed or transmitted.
[0003]
[Patent Document 1]
JP2001-256256
[0004]
[Problems to be solved by the invention]
In Patent Document 1, when it is desired to search and print an original document, even if the degree of similarity is sufficiently large, the search result is once displayed on the display unit, and printing or transmission is selected, which requires extra work. .
[0005]
[Means for Solving the Problems]
In order to solve the above-mentioned problem, an image processing apparatus according to claim 1 of the present invention is an image processing apparatus that searches an image similar to an input document image from registered data. An input means for inputting a document image; Above Entered by input means As a result of the calculation by the similarity calculation means for calculating the similarity between the document image and the registered data, and the similarity calculation means, the similarity with the document image is Most Registration data that has been determined that the difference between the similarity of the highest registered data and the similarity of the next highest registered data is greater than a predetermined value Address The A notification means for automatically notification, and registered data whose address is notified by the notification means. Print control means for controlling printing, and
It is characterized by having.
[0006]
In order to solve the above problem, an image processing method according to claim 4 of the present invention is an image processing method for retrieving an image similar to an input document image from registered data. An input step of inputting a document image by an input means; Above Entered by input means The similarity between the document image and the registered data Similarity calculation means The difference between the similarity calculation step to be calculated and the similarity of the registration data having the highest similarity to the document image and the similarity of the registration data having the next highest similarity is predetermined as a result of the calculation by the similarity calculation step. Registration data determined to be greater than the value of Address The A notification step automatically notified by the notification means, and registration data for which the address was notified in the notification step To print Print control means And a printing control step for controlling.
[0010]
DETAILED DESCRIPTION OF THE INVENTION
Embodiments of the present invention will be described. FIG. 1 is a block diagram showing a configuration example of an image processing apparatus according to the present invention. In the present embodiment, an environment in which the office 10 and the office 20 are connected by the Internet 104 is given. A LAN 107 constructed in the office 10 is connected to the MFP 100, a management PC 101 that controls the MFP 100, a client PC (external storage means) 102, a document management server 106, its database 105, and a proxy server 103. The LAN 107 and the LAN 108 in the office 20 are connected to the Internet 104 via the proxy server 13. In the present invention, the MFP 100 is in charge of a part of image processing for the image signal read by the image reading unit of the paper document, and the image signal is input to the management PC 101 using the LAN 109. The management PC is a normal PC and includes an image storage unit, an image processing unit, a display unit, and an input unit. A part of the management PC is integrated with the MFP 100.
[0011]
FIG. 2 is a configuration diagram of the MFP 100. In FIG. 2, an image reading unit 110 including an auto document feeder (hereinafter referred to as ADF) irradiates a bundle or one original image with a light source (not shown), and forms an original reflection image on a solid-state image sensor with a lens. A raster-like image reading signal is obtained as image information from the solid-state imaging device. In the normal copying function, this image signal is image-processed into a recording signal by the data processing unit 115. In the case of copying every plural number, recording data for one page is temporarily stored in the recording unit 111 and then sequentially output to the recording unit 112. Then, an image is formed on the paper.
[0012]
On the other hand, print data output from the client PC 102 is converted into raster data that can be recorded by the data processing unit 115 from the LAN 107 via the network IF 114, and then formed as a recorded image on paper by the recording unit.
[0013]
An operator's instruction to the MFP 100 is performed from a key operation unit provided in the MFP and an input unit 113 including a keyboard and a mouse that are input to the management PC. These series of operations are a control unit (not shown) in the data processing unit 115. It is controlled by.
[0014]
On the other hand, the status display of the operation input and the display of the image data being processed are performed on the display unit 116. The storage unit 111 is also controlled by the management PC, and data exchange and control between the MFP and the management PC are performed using the network IF 117 and the directly connected LAN 109.
[0015]
〔Outline of processing〕
Next, an outline of image processing according to the present invention will be described with reference to FIG.
[0016]
In document input processing (2001) for inputting a document, the image reading unit 110 of the MFP 100 is operated to scan one document in a raster pattern to obtain an image signal. Next, in the determination process (2002) for determining the process set in advance in the process setting, the setting set in the user interface as shown in FIG. 6 is determined. If document output is set, the image input in 2001 is output as it is to print / edit / store / transmit / record the image (2004). If an original output for searching for an original is set, an original process (2003) is performed and output to print / edit / store / transmit / record the image (2004).
[0017]
[Outline of original processing]
Next, an outline of the original processing of image processing according to the present invention will be described with reference to FIG.
[0018]
The image signal input in the document input process is preprocessed by the data processing unit 115 and stored in the storage unit 111 as image data for one page. The CPU of the management PC 101 first separates the area from the stored image signal into a character / line image portion and a halftone image portion, and the character portion is further divided into blocks or a block or a line. Separated into organized tables and figures and segmented. On the other hand, the image portion expressed by halftone is divided into independent objects for each so-called block, such as an image portion of a block separated into rectangles, a background portion, and the like (step 121).
[0019]
At this time, a two-dimensional barcode recorded as additional information in the original image or an object corresponding to the URL is detected, and the URL recognizes the character by OCR. If the two-dimensional barcode is used, the mark is decoded (step 122). Pointer information in the storage unit storing the original electronic file of the original is detected (step 123). There are other means for adding pointer information, such as a method of embedding information in the space between characters and a method of embedding in a halftone image, such as a so-called digital watermark method that is not directly visualized.
[0020]
If the pointer information is detected in step 124, the process branches to step 125, and the original electronic file is searched from the address indicated by the pointer. The electronic file is a document registered by scanning, a document created by an application, or the like, and is stored in the hard disk in the client PC in FIG. 1 or in the document management server 105 connected to the LAN of the office 10 or 20. It is stored either in the database 105 or in the storage unit 111 included in the MFP 100 itself. If the electronic file is not found in step 125, if it is found but is a so-called image file represented by PDF or TIFF, or if the pointer information itself does not exist in step 124, the process branches to step 126.
[0021]
In step 126, in order to search the original electronic file on the database, first, the input image is converted into vector data. First, for the character block that has been OCR in step 122, the character size, style, and font are further recognized and converted into font data that is visually faithful to the character obtained by scanning the document. A table or figure block composed of one line is outlined, and if the figure shape such as a table can be recognized, the shape is recognized. The image block is processed as an individual image file as image data. These vectorization processes are performed for each object. The image converted to the file vector data on the database is checked for similarity with each file on the database in step 127, and the original is searched. In this embodiment, the original file is searched faithfully using the vector data converted in step 126. The degree of similarity is obtained for each object, and the degree of similarity for each object is reflected in the degree of similarity of the entire file according to the occupancy ratio of the object in the file. Since the degree of similarity of objects that occupy a large percentage in the file is more greatly reflected in the degree of similarity of the entire file, it is possible to adaptively handle files of any format.
[0022]
As a result of comparing the similarity with the threshold value in step 128, when the candidate is one file, the similarity of the file is compared with the predetermined threshold value when the number of candidates is plural and the similarity of the file with the highest similarity is compared. If it is higher than the threshold value, the process automatically branches to step 134 to notify the storage address. Note that this branch determination does not compare with the threshold value, but as a branch condition for branching to 134 if the difference between the highest similarity and the second highest similarity is greater than a predetermined value. Alternatively, it may be configured such that the file having the highest similarity is selected unconditionally and the process proceeds to step 134 without setting a branch. By not interposing a user's selection operation between scanning and receiving an output such as printing, operability can be greatly improved.
[0023]
If there is no file whose similarity exceeds the threshold value in step 128, thumbnails and the like are displayed in the order of similarity as shown in FIG. 7 (step 129). Specify a file from among the files. In step 130, there is an electronic file desired by the user in the file displayed in step 129, and if it is selected, the process branches to step 134 to notify the storage address of the file. Branch to
[0024]
In step 131, vectorization processing is performed to register the input data. The vectorization process is performed for each object, and further, layout information of each object is stored, converted into, for example, rtf (step 131), and stored as an electronic file in the storage unit 111 (step 132).
[0025]
First, in step 133, index information for search is generated and added to the search index file so that the vectorized document image can be directly searched as an electronic file when the same processing is performed thereafter. In step 134, the storage address when it is stored in the storage unit is notified to the operator.
[0026]
As described above, using the electronic file itself obtained by the present invention, for example, printing, transmission, processing, storage, and recording of a document can be performed in step 135. In the above embodiment, the storage address is notified to the operator. However, the document may be printed, transmitted, processed, stored, and recorded without notification.
[0027]
Hereinafter, each processing block will be described in detail.
[0028]
First, the block selection process shown in step 121 will be described.
[0029]
[Block selection processing]
As shown in FIG. 4, the block selection process recognizes a document image as a block for each object, determines each block as an attribute such as character / drawing / photo / line / table, etc. This is a process of dividing.
[0030]
An example of the block selection process will be described below.
[0031]
First, the input image is binarized into black and white, and contour tracking is performed to extract a block of pixels surrounded by a black pixel contour. For a black pixel block with a large area, contour tracing is also performed for white pixels inside, and a white pixel block is extracted, and the black pixel block is recursively extracted from the white pixel block with a certain area or more. Extract lumps.
[0032]
The black pixel blocks obtained in this way are classified by size and shape, and are classified into regions having different attributes. For example, if the aspect ratio is close to 1 and the size is within a certain range, the pixel block corresponding to the character is used, the portion where the adjacent characters can be grouped in a well-aligned manner is the character region, and the flat pixel block is the line region. The area occupied by the black pixel block that is larger than the size and contains the square white pixel block in a well-aligned manner is the table region, the region where the irregular pixel block is scattered is the photo region, and the pixel block of any other shape is used. A drawing area, etc.
[0033]
FIG. 4 shows block information for each block obtained by the block selection process.
[0034]
Information for each block is used as information for vectorization or search described below.
[0035]
[Character recognition]
The character recognition unit recognizes an image cut out in character units using a pattern matching method, and obtains a corresponding character code. This recognition process recognizes the character type with the closest distance by comparing the observed feature vector obtained by converting the feature obtained from the character image into a numerical sequence of several tens of dimensions and the dictionary feature vector obtained for each character type in advance. The resulting process. There are various known methods for extracting a feature vector. For example, there is a method characterized by dividing a character into meshes, and using a mesh number-dimensional vector obtained by counting character lines in each mesh as line elements according to directions.
[0036]
When character recognition is performed on the character area extracted in block selection (step 121), first, horizontal writing and vertical writing are determined for the corresponding area, lines are cut out in the corresponding directions, and then characters are cut out. Get a character image. Horizontal / vertical writing can be determined by taking a horizontal / vertical projection of the pixel value in the corresponding area, and determining that the horizontal projection area is large when the horizontal projection variance is large, and vertical writing area when the vertical projection variance is large. . For horizontal writing, character strings and characters are decomposed by cutting out lines using horizontal projection, and then cutting out characters from the vertical projection of the cut lines. For vertically written character areas, horizontal and vertical may be reversed. At this time, the character size can be detected.
[0037]
[File Search]
Next, details of the file search process shown in step 127 of FIG. 3 will be described with reference to FIGS.
[0038]
In this embodiment, a search is performed using each block information divided and vectorized by the block selection process described above. Specifically, the search is a comparison between the attribute of each block and the block coordinate information in the file, that is, comparison by layout, and internal information comparison for each block to which a different comparison method is applied depending on the attribute of each block in the file. Use complex compound search.
[0039]
FIG. 11 shows an example of the scanned image data (input file) vectorized at step 126 of FIG. 3, which is divided into blocks B′1 to B′9 and each vectorized.
[0040]
FIG. 12 is a flowchart for calculating the similarity by sequentially comparing the input file with image data (database file) already vectorized and stored on the database. First, the database file is accessed from the database (step 501). Each block of the input file is compared with each block of the database file, and a similarity with the block of the database file is obtained for each block of the input file (step 502).
[0041]
Here, when calculating the similarity for each block, the target block of the database file that is presumed to be identical in layout to the block of the input file is first selected according to the flowchart shown in FIG. In this process, the target block of the database file may be selected in duplicate for a plurality of blocks of the input file. Next, the similarity of the layout information between the block and the target block is obtained. The block positions, sizes, and attributes are compared (steps 512, 513, and 514), and the layout similarity is obtained from the error. Next, the inside of the block is compared. When the inside of the block is compared, the same attribute is compared. Therefore, if the attributes are different, pre-processing such as re-vectorizing one block to the matching attribute is performed. The block of the input file and the target block of the database file that are treated as having the same attribute by the preprocessing are subjected to block internal comparison (step 515).
[0042]
In the block internal comparison, since the optimum comparison method is adopted for the block attribute, the comparison method differs depending on the attribute. For example, the block is divided into attributes such as text, photograph, table, and line drawing by the block selection process described above. When comparing text blocks, since the character code and font are determined by vectorization processing, the similarity of the sentence is calculated from the matching degree of each character, and the similarity inside the block is calculated. In the photographic image block, the similarity is calculated from the feature vector extracted from the image based on the error in the feature space. The feature vector here includes a plurality of color feature values such as a color histogram and a color moment, a co-occurrence matrix, a texture feature amount expressed by contrast, entropy, Gabor transformation, and a shape feature amount such as a Fourier descriptor. The optimum combination is used among the plurality of feature quantities. Also, in a line drawing block, the line drawing block is expressed as an outline line, a ruled line, or a set of curves by vectorization processing. Therefore, by calculating errors such as the start point, end point position, and curvature of each line, the degree of similarity of line drawing can be calculated Calculated. In the table block, the similarity of the entire table block can be calculated by calculating the number of grids in the table, the size of each frame, the text similarity in each grid, and the like.
[0043]
From the above, it is possible to calculate the block position, size, attribute, similarity inside the block, and to calculate the similarity for the block of the input file by summing up the similarities. Record. A series of processing is repeated for all blocks of the input file. All of the obtained block similarities are integrated to obtain the similarity of the input file (step 503). The integration process will be described. Assume that the similarity for each block is calculated as n1 to n9 for the blocks B1 ′ to B9 ′ of the input file in FIG. At this time, the overall similarity N of the entire file is expressed by the following equation.
N = w1 * n1 + w2 * n2 + w3 * n3 +. + W9 * n9 + γ (1)
[0044]
Here, w1 to w9 are weighting factors for evaluating the similarity of each block. γ is a correction term, for example, an evaluation value of a block not selected as a target block for the input file of the database file. The weighting factors w1 to w9 are obtained from the occupancy rate of the block in the file. For example, if the sizes of the blocks 1 to 9 are S1 to S9, the occupation ratio w1 of the block 1 is
w1 = S1 / (S1 + S2 +... + S9) (2)
Can be calculated as By such weighting processing using the occupation ratio, the similarity of blocks that occupy a large area in the file is more reflected in the similarity of the entire file.
[0045]
[Calculating text search similarity in file search]
When a document is registered, words included in the registered document are acquired. Next, it is calculated from words appearing in the document using a basic vector dictionary. FIG. 9 shows the structure of the basic vector dictionary. The basic vector dictionary stores a feature quantity corresponding to each dimension (Dim.) At the time of vector expression for each word. For the dimension, a standard classified according to the original meaning of the word, a standard classified according to the field of use of the word, or the like is adopted. Dim. 1 is 0, and Dim. It can be seen that the feature amount of 2 is 23. In this way, it is possible to obtain the feature quantity of each dimension (Dim.) In one word from the dictionary. The feature amount can be interpreted as a value indicating how much the document may characterize the classification standard (= dimension) by using the word. The feature amount of the entire document is expressed by a vector having the classification reference as a dimension from the feature amounts of the classification reference (by dimension) obtained from all the words constituting the document. A value obtained by normalizing the obtained vector with norm = 1 is stored as a document vector. The document vector is stored in an index as shown in FIG. Dim. Of document vector of document ID = 6947. 1 is 0.183, and Dim. It can be seen that the feature amount of 2 is 0.214.
[0046]
[Conversion to application data]
By the way, the image data for one page is subjected to block selection processing (step 121), and the result of vectorization processing (step 126) is converted as an intermediate data format file as shown in FIG. The format is called Document Analysis Output Format (DAOF).
[0047]
FIG. 14 shows the data structure of DAOF.
[0048]
In FIG. 14, reference numeral 791 denotes a header, which holds information relating to document image data to be processed. In the layout description data portion 792, attributes such as TEXT (character), TITLE (title), CAPTION (caption), LINEART (line drawing), EPICTURE (natural image), FRAME (frame), TABLE (table), etc. in the document image data. The attribute information of each block recognized every time and its rectangular address information are held. The character recognition description data portion 793 holds character recognition results obtained by character recognition of TEXT blocks such as TEXT, TITLE, and CAPTION. The table description data portion 794 stores details of the structure of the TABLE block. The image description data portion 795 cuts out image data of blocks such as PICTURE and LINEART from the document image data and holds them.
[0049]
Such a DAOF is not only used as intermediate data but may be stored as a file itself. In this file state, it is not possible to reuse individual objects in a so-called general document creation application. I can't. Next, the process of converting DAOF to application data (step 131) will be described in detail.
[0050]
FIG. 15 is a schematic flow of application data conversion.
In 8000, DAOF data is input.
8002 generates a document structure tree that is the source of application data.
In step 8004, actual data in the DAOF is poured based on the document structure tree to generate actual application data.
[0051]
FIG. 16 is a detailed flow of the 8002 document structure tree generation unit, and FIG. 17 is an explanatory diagram of the document structure tree. As a basic rule of overall control, the flow of processing shifts from a micro block (single block) to a macro block (an aggregate of blocks).
[0052]
Hereinafter, the block refers to the micro block and the entire macro block.
[0053]
8100 performs regrouping based on the relevance in the vertical direction in units of blocks. Immediately after the start, judgment is made in units of micro blocks.
[0054]
Here, the relevance can be defined by the fact that the distance is close and the block width (height in the horizontal direction) is substantially the same.
[0055]
Information such as distance, width, and height is extracted with reference to DAOF.
[0056]
FIG. 17A shows an actual page configuration, and FIG. 17B shows its document structure tree. As a result of 8100, T3, T4, and T5 are generated as one group V1, and T6 and T7 are generated as one group V2 in the same hierarchy.
[0057]
8102 checks whether or not there is a separator in the vertical direction. For example, the separator is physically an object having a line attribute in the DAOF. Also, logically, it is an element that explicitly divides a block in the application. If a separator is detected here, it is subdivided at the same level.
[0058]
8104 uses the group length to determine whether there are no more divisions.
[0059]
If the group length in the vertical direction is the page height, the document structure tree generation ends.
[0060]
In the case of FIG. 17, since there is no separator and the group height is not the page height, the process proceeds to 8106.
[0061]
In step 8106, regrouping is performed based on the relevance in the horizontal direction in units of blocks. Here too, the first time immediately after the start is determined in units of microblocks.
[0062]
The definition of the relevance and the determination information is the same as in the vertical direction.
[0063]
In the case of FIG. 17, H1 at T1 and T2 and H2 at V1 and V2 are generated as a group in the same hierarchy one above V1 and V2.
[0064]
8108 checks for the presence of a horizontal separator.
[0065]
In FIG. 17, since there is S1, this is registered in the tree, and a hierarchy of H1, S1, and H2 is generated.
[0066]
8110 uses the group length to determine whether there are no more divisions.
[0067]
If the horizontal group length is the page width, the document structure tree generation ends.
[0068]
If not, the process returns to 8102, and the relevance check in the vertical direction is repeated again at the next higher level.
[0069]
In the case of FIG. 17, since the division width is the page width, the process ends here, and finally V0 of the highest hierarchy representing the entire page is added to the document structure tree.
[0070]
After the document structure tree is completed, application data is generated in 8006 based on the information.
[0071]
Specifically, in the case of FIG.
[0072]
That is, since there are two blocks T1 and T2 in the horizontal direction, H1 has two columns, and after T1 internal information (refer to DAOF, text of character recognition result, image, etc.) is output, the column is changed and the internal of T2 Information is output, and then S1 is output.
[0073]
Since H2 has two blocks V1 and V2 in the horizontal direction, it outputs as two columns, V1 outputs its internal information in the order of T3, T4, T5, then changes the column, and outputs the internal information of T6, T7 of V2 To do.
[0074]
As described above, conversion processing to application data can be performed.
[0075]
[Another Example 1 in File Search]
In the above embodiment, when comparing the input file and the database file in the file search, the layout information and the internal information of the block are compared for all the blocks of all the input files. However, it is possible to select files to some extent at the stage of comparing layout information without comparing block internal information. In other words, a database file whose layout is completely different from that of the input file can omit the block internal information comparison process. FIG. 19 is a flowchart when file selection is performed based on layout information. First, the positions, sizes, and attributes of all the blocks of the input file are compared, the similarity is obtained, and the layout similarity of the entire file is obtained (step 522). If the layout similarity is lower than the threshold, the block internal information comparison is not performed (step 523). Only when it is higher than the threshold value, that is, when the layout is similar, the block internal information is compared (step 524), and the overall similarity of the entire file is obtained from the previously obtained layout similarity and the similarity inside the block ( Step 525). The method for obtaining the total similarity from the similarity for each block is the same processing as step 503 in FIG. A file whose similarity is equal to or higher than a threshold is stored as a candidate. The process for obtaining the similarity of the block internal information is generally a heavy process, particularly when checking the match of a photo block. Therefore, by narrowing down the files to some extent by the layout, the amount of search processing can be reduced and the processing speed can be increased, and a desired file can be searched efficiently.
[0076]
[Another Example 2 in File Search]
All of the above-described embodiments are search processing embodiments when a user performs a search without specifying anything when searching for a file. However, it is possible to further optimize the file search by allowing the user to specify a part to be a feature in the document (a block obtained from block selection), omit a useless block, or specify a feature in the document. It becomes possible.
[0077]
FIG. 8 shows an example of a user interface screen (1001) for specifying a search option by the user at the time of search. The input file is divided into a plurality of blocks by block selection processing, and each block such as text, photo, table, and line drawing on the file is displayed as a thumbnail on the input screen (1011 to 1017). The user selects a block as a feature from the displayed blocks. A plurality of blocks may be selected at this time. As an example, assume that block 1014 is selected. When the button “priority” (1003) is pressed in a state where the block 1014 is selected, a search process that places more importance on the block 1014 is performed. The important search is, for example, to increase the weight coefficient of the block 1014 designated in the calculation formula (1) for obtaining the similarity of the entire file from the similarity of each block, and to decrease the weight coefficient of the non-selected block. This can be achieved. If the “priority” button (1004) is pressed a plurality of times, the weighting coefficient of the selected block is increased, and a search with more emphasis on the block can be performed. If the exclude button (1004) is pressed, the search process is performed with the selected block 1014 omitted. When a block is recognized by mistake, useless search processing can be omitted and erroneous search processing can be prevented. Further, it is possible to change the attribute of the block by the detailed setting (1005), and even when the attribute is mistakenly recognized in the block selection (step 121), the user can correct it and correct the search. Also, in the detailed setting 1005, the user can finely adjust the weight for prioritizing the block search. As described above, when a search is performed, the user can specify and set a characteristic block to optimize the search.
[0078]
On the other hand, some files may have special layouts. For such a file, selecting the layout priority button (1005) in FIG. 8 enables file search with an emphasis on layout. In this case, weighting is performed so that the layout similarity result is more important. The text priority button (1006) can search only text blocks and reduce processing.
[0079]
In this way, by allowing the user to select image features, a search that emphasizes file features can be performed. In addition, when the user's artificial means is trusted, that is, when the weight is changed by the user designation, if the restriction that the changed weight becomes less than the threshold value is not searched, the restriction of the user is simplified. It is also possible to significantly reduce useless block search processing with simple operations.
[0080]
(Other examples)
In the above embodiment, as shown in FIG. 6, processing is selected and executed from original output and original output, but the present invention is not limited to this. As shown in FIG. 20, processing may be selected from original output, original registration, original output (original registration is not performed), and original output (original registration is performed). If original registration is selected, the process starts from step 131 shown in FIG. 3 after inputting the image, and the image is not printed. If document output (original registration is not registered) is selected, after inputting the image, the process jumps to step 135 to print the image. If document output (original registration) is selected, the process starts from step 131 after image input, and registration and image printing are performed.
[0081]
In the above embodiment, the threshold value and setting value to be compared in step 128 are set in advance, but means for setting the threshold value may be provided. In that case, for example, the setting may be made with an interface as shown in FIG.
[0082]
【The invention's effect】
As described above in detail, according to the present invention, in an image processing apparatus, registration data having a high degree of similarity between an input image and registration data is printed without the user's selection operation, thereby greatly improving user operability. It becomes possible to improve.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of a system according to an embodiment of the present invention.
FIG. 2 is a block diagram showing a configuration of an MFP according to the embodiment of the present invention.
FIG. 3 is a flowchart showing an original processing procedure according to the embodiment of the present invention.
FIG. 4 is an example of block selection processing according to the embodiment of the present invention.
FIG. 5 is a flowchart showing a schematic processing procedure;
FIG. 6 is a diagram illustrating an example of a user interface screen.
FIG. 7 is a diagram illustrating an example of a list selection user interface screen.
FIG. 8 is a diagram illustrating an example of a user interface screen.
FIG. 9 is an example of a basic vector dictionary for text search.
FIG. 10 is an example of a text document vector index.
FIG. 11 is a diagram illustrating a block example.
FIG. 12 is a flowchart showing a processing procedure for file search processing;
FIG. 13 is a flowchart illustrating a block comparison process procedure of a file search process.
FIG. 14 is a diagram illustrating an example of a DAOF.
FIG. 15 is a flowchart showing an application data conversion processing procedure.
FIG. 16 is a flowchart showing a document structure tree generation processing procedure;
FIG. 17 is an explanatory diagram of a document structure tree.
FIG. 18 is a diagram illustrating an example of a threshold setting user interface screen.
FIG. 19 is a flowchart showing a file selection processing procedure based on layout information.
FIG. 20 is a diagram illustrating an example of a user interface screen.

Claims (5)

In an image processing apparatus for retrieving an image similar to an input document image from registered data,
An input means for inputting a document image;
Similarity calculating means for calculating the similarity between the document image input by the input means and registered data;
The calculation by the similarity calculating unit results, similarity between the document image is determined as the difference between the similarity of the next degree of similarity is high registration data similarity of the most high registration data is larger than a predetermined value Notification means for automatically notifying the address of the registered data ,
Print control means for controlling to print registration data whose address is notified by the notification means ;
An image processing apparatus comprising:   If the difference between the similarity of the registered data having the highest similarity to the document image and the similarity of the registered data having the next highest similarity is not greater than the predetermined value, the plurality of registered data are set to the similarity. The image processing apparatus according to claim 1, further comprising list display means for displaying a list based on the list.   The image processing apparatus according to claim 1, further comprising a value setting unit configured to set the predetermined value. In an image processing method for retrieving an image similar to an input document image from registered data,
An input step of inputting a document image by an input means;
A similarity calculating step in which the similarity calculating means calculates the similarity between the document image input by the input means and the registered data;
As a result of the calculation by the similarity calculation step, it is determined that the difference between the similarity of the registered data having the highest similarity with the document image and the similarity of the registered data having the next highest similarity is larger than a predetermined value. A notification step in which the notification means automatically notifies the address of the registered data ;
A print control step in which the print control means controls to print the registration data notified of the address in the notification step;
An image processing method comprising:   A computer program comprising program code for causing a computer to execute each step described in the image processing method according to claim 4.

Images