CN105160343A - Information identification method and device applied to film on-demand-printing system - Google Patents
Information identification method and device applied to film on-demand-printing system Download PDFInfo
- Publication number
- CN105160343A CN105160343A CN201510491352.6A CN201510491352A CN105160343A CN 105160343 A CN105160343 A CN 105160343A CN 201510491352 A CN201510491352 A CN 201510491352A CN 105160343 A CN105160343 A CN 105160343A
- Authority
- CN
- China
- Prior art keywords
- film
- character
- information
- identified
- demand system
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/32—Normalisation of the pattern dimensions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
- G06V30/148—Segmentation of character regions
- G06V30/153—Segmentation of character regions using recognition of characters or words
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/28—Character recognition specially adapted to the type of the alphabet, e.g. Latin alphabet
- G06V30/293—Character recognition specially adapted to the type of the alphabet, e.g. Latin alphabet of characters other than Kanji, Hiragana or Katakana
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Character Input (AREA)
Abstract
The invention discloses an information identification method applied to a film on-demand-printing system. The method comprises steps of receiving a to-be-identified film uploaded by the film on-demand-printing system; positioning an information identification area in the to-be-identified film according to a preset digital template; performing image processing of the to-be-identified film and segmenting characters in the positioned information identification area; and identifying a single character obtained after character segmentation, and determining identification information in the information identification area according to the identified single character. The invention also discloses a corresponding device. The invention combines characteristics of image information of films generated by equipment work stations and diagnosis typesetting work stations, solves problems that the conventional film on-demand-printing system is poor in identification effect for electronic films generated by different pieces of equipment or different layout films generated by the same equipment, is wide in application range and high in film information identification efficiency, and achieves automatic and accurate matching between the film and the corresponding patient.
Description
Technical field
The present invention relates to information discriminating technology field, particularly relate to a kind of information identifying method and the device that are applied to film print on demand system.
Background technology
The patient information identification module of film print on demand system traditional at present, normally feature extraction and identification are carried out to entire image, but this recognition methods is of limited application due to it, not high to the discrimination of the electronic image memory device of the different imposition layouts that image documentation equipment generates, algorithm complex is high, and result in electronic image memory device cannot with corresponding patient automatically, the problem of accurate match; Further, when the film distribution that equipment workstation produces is too much, recognition efficiency can be lower.
Summary of the invention
Fundamental purpose of the present invention is to provide a kind of information identifying method being applied to film print on demand system, applied range, high to the information recognition efficiency of film, and can realize automatic, the accurate match of film and corresponding patient information.
For achieving the above object, the invention provides a kind of information identifying method being applied to film print on demand system, described in be applied to film print on demand system information identifying method comprise the steps:
Receive the film to be identified that film print on demand system is uploaded;
Digital template according to presetting positions the information identified region in described film to be identified;
Image procossing is carried out to described film to be identified, and Character segmentation is carried out to the character in the described information identified region of location;
The single character obtained after identification character segmentation, determines the identifying information in described information identified region according to the described single character identified.
Preferably, before the step of the film to be identified that described reception film print on demand system is uploaded, described in be applied to film print on demand system information identifying method also comprise:
There is provided digital template to make interface, the identification parameter for the film to be identified generated according to each film print on demand system arranges digital template;
Digital template described in each of preservation setting.
Preferably, the digital template that described basis is preset comprises the step that the information identified region in described film to be identified positions:
Digital template described in each is read in circulation;
The reference coordinate stored in the coordinate of the information identified region in described film to be identified and described digital template is contrasted, determines the information identified region in film to be identified.
Preferably, described image procossing is carried out to film to be identified, and the step that the character in the described information identified region of location carries out Character segmentation is comprised:
Binary conversion treatment is carried out to the image of described film to be identified;
Determine the upper and lower border of the character in the information identified region of the described film after binary conversion treatment and left and right border, read single characters all in this information identified region.
Preferably, the single character obtained after the segmentation of described identification character, the step of the identifying information determined in described information identified region according to the described single character identified comprises:
Carry out character normalization process to the single character obtained after Character segmentation, described character normalization process comprises character position normalized and character boundary normalized;
Reference character pixel in the pixel of the described single character after contrast character normalization process and described digital template, identifies whole single characters;
The whole described single character of linkage identification, obtains the identifying information in described information identified region.
In addition, the present invention also provides a kind of information recognition device being applied to film print on demand system, described in be applied to film print on demand system information recognition device comprise:
Receiver module, for receiving the film to be identified that film print on demand system is uploaded;
Locating module, for positioning the information identified region in described film to be identified according to the digital template preset;
Process and segmentation module, for carrying out image procossing to described film to be identified, and carry out Character segmentation to the character in the described information identified region of location;
Identify and determination module, for the single character obtained after identification character segmentation, determine the identifying information in described information identified region according to the described single character identified.
Preferably, the information recognition device being applied to film print on demand system described in also comprises:
Interface provides module, and for providing digital template to make interface, the identification parameter for the film to be identified generated according to each film print on demand system arranges digital template;
Preserve module, for preserve setting each described in digital template.
Preferably, described locating module specifically for:
Digital template described in each is read in circulation;
The reference coordinate stored in the coordinate of the information identified region in described film to be identified and described digital template is contrasted, determines the information identified region in film to be identified.
Preferably, described process and segmentation module specifically for:
Binary conversion treatment is carried out to the image of described film to be identified;
Determine the upper and lower border of the character in the information identified region of the described film after binary conversion treatment and left and right border, read single characters all in this information identified region.
Preferably, described identification and determination module specifically for:
Carry out character normalization process to the single character obtained after Character segmentation, described character normalization process comprises character position normalized and character boundary normalized;
Reference character pixel in the pixel of the described single character after contrast character normalization process and described digital template, identifies whole single characters;
The whole described single character of linkage identification, obtains the identifying information in described information identified region.
The present invention is after receiving the film to be identified that film print on demand system uploads, digital template according to presetting positions the information identified region in film to be identified, image procossing is carried out to film to be identified, and Character segmentation is carried out to the character in the information identified region of location; The single character obtained after identification character segmentation, according to the identifying information in the single character comformed information identified region identified.The feature of the film image information that embodiment of the present invention bonding apparatus workstation and diagnosis typesetting workstation produce, solve electronic image memory device that current film print on demand system produces for distinct device or the not good problem of the different layout film image recognition effects that same equipment produces, its applied range, high to the information recognition efficiency of film, and automatic, the accurate match of film and corresponding patient information can be realized.
Accompanying drawing explanation
Fig. 1 is the schematic flow sheet that the present invention is applied to information identifying method first embodiment of film print on demand system;
Fig. 2 is the schematic flow sheet that the present invention is applied to information identifying method second embodiment of film print on demand system;
Fig. 3 is the refinement schematic flow sheet of step S20 in Fig. 1;
Fig. 3 a-Fig. 3 d is the display schematic diagram that digital template makes interface;
Fig. 4 is the refinement schematic flow sheet of step S30 in Fig. 1;
Fig. 4 a and Fig. 4 b is the upper and lower border of character in information identified region and the schematic diagram on left and right border;
Fig. 5 is the refinement schematic flow sheet of step S40 in Fig. 1;
Fig. 6 is the high-level schematic functional block diagram that the present invention is applied to information recognition device first embodiment of film print on demand system;
Fig. 7 is the high-level schematic functional block diagram that the present invention is applied to information recognition device second embodiment of film print on demand system.
The realization of the object of the invention, functional characteristics and advantage will in conjunction with the embodiments, are described further with reference to accompanying drawing.
Embodiment
Should be appreciated that specific embodiment described herein only in order to explain the present invention, be not intended to limit the present invention.
The invention provides a kind of information identifying method being applied to film print on demand system.
Reference Fig. 1, Fig. 1 are the schematic flow sheet that the present invention is applied to information identifying method first embodiment of film print on demand system.
In a first embodiment, the information identifying method being applied to film print on demand system comprises:
Step S10, receives the film to be identified that film print on demand system is uploaded;
Step S20, the digital template according to presetting positions the information identified region in film to be identified;
The embodiment of the present invention can be applicable to various film print on demand system, and before identifying the identifying information in film to be identified, be provided with corresponding digital template for the issuable film of various different film print on demand systems and store, to be identified by the identifying information of digital template to the film to be identified received.After receiving the film to be identified that film print on demand system uploads, digital template according to presetting positions the information identified region in film to be identified, each digital template that concrete localization method stores by traversal, information identified region and digital template are compared, finally determines the position at the information identified region place in film to be identified.
Step S30, carries out image procossing to film to be identified, and carries out Character segmentation to the character in the information identified region of location;
Step S40, the single character obtained after identification character segmentation, according to the identifying information in the single character comformed information identified region identified.
Determine the information identified region in film to be identified, in order to the character in follow-up identifying information identified region, thus image procossing is carried out to this film to be identified, and further Character segmentation is carried out to the character in the information identified region of location, then the single character obtained after Character segmentation is identified one by one, and according to the identifying information in the final comformed information identified region of each single character identified, thus obtain the relevant identification information in pending film.
The present embodiment is after receiving the film to be identified that film print on demand system uploads, digital template according to presetting positions the information identified region in film to be identified, image procossing is carried out to film to be identified, and Character segmentation is carried out to the character in the information identified region of location; The single character obtained after identification character segmentation, according to the identifying information in the single character comformed information identified region identified.The feature of the film image information that embodiment of the present invention bonding apparatus workstation and diagnosis typesetting workstation produce, solve electronic image memory device that current film print on demand system produces for distinct device or the not good problem of the different layout film image recognition effects that same equipment produces, its applied range, high to the information recognition efficiency of film, and automatic, the accurate match of film and corresponding patient information can be realized.
Reference Fig. 2, Fig. 2 are the schematic flow sheet that the present invention is applied to information identifying method second embodiment of film print on demand system.
Based on above-described embodiment, propose the present invention and be applied in information identifying method second embodiment of film print on demand system, before step S10, the method also comprises:
Step S50, provide digital template to make interface, the identification parameter for the film to be identified generated according to each film print on demand system arranges digital template;
Step S51, preserves each digital template arranged.
Due to different image documentation equipments produce printing film identifying information residing for position different, for the electronic image memory device of the different printing layouts that same image documentation equipment produces, position residing for identifying information is also not identical usually, moreover, identification character morphological feature, identify length also by may be different.Therefore, before identification film, digital template is provided to make interface to user, the printing film produced for various different image documentation equipment by this digital template making interface for user makes digital template, while making digital template, record the position that in the film of each printing layout, identifying information is corresponding, namely all make set of number template and corresponding identification parameter for the printing film of different images equipment generation and the multiple layout film of same equipment.Each equipment workstation is sent in the configuration information of virtual printing service end, all comprise the information of an AETitle, this information unique identification sends the equipment workstation model of image, is determined the source device of this image before carrying out image identification by the AETitle item read in configuration information.
Before whole print on demand system drops into formal use, need to carry out the making of respective digital template and the record of configuration information to all layouts of each equipment of hospital come into operation, therefore before use system, need to carry out alternately with corresponding hospital, the film data of the various layouts that the image documentation equipment workstation commonly used to obtain them generates.Film data are normally transmitted with the form of DICOM file, and DICOM file is a kind of special file, not only contains image, and it also comprises the basic parameter of the electronic image memory device correction of image, as image width, highly, picture element density etc.After DICOM file is presented at digital template making interface, need to fill in the partial parameters information at interface, current partial image equipment, the information identified and patient ID or checking sequence number may be not only digital, but the combination of numeral and letter, the situation of letter and number cross arrangement seldom occurs, therefore, the feature little in conjunction with alphabetical probability of occurrence and equipment generation film image letter is identical, make in interface in digital template and prefix and suffix information are recorded, if identifying information is without prefix and suffix, correspondence position puts sky, identify the rear corresponding informance reading record, final recognition result is obtained by combination.
Simultaneously, digital template makes interface based on the message response function in MFC, and by amendment response function code, instrument is supported, and mouse frame favored area is converted to the function of coordinate, by recording the position coordinates of identified region after conversion, and save as follow-up use.After frame selects identified region to complete, the discriminating digit that input frame favored area is corresponding, and electronic image memory device that selection this time makes is which layout sequence of this equipment, only has current layout of choosing to be greater than just permission when there is layout.After the essential information of electronic image memory device necessity completes, generate Template button and complete the current digital template chosen by clicking and generate.Need when template construct completes to preserve the film configuration information of this layout, utilize the function defined in LiniFileConfig class to preserve configuration information.
With reference to the refinement schematic flow sheet that Fig. 3, Fig. 3 are step S20 in Fig. 1.
Be applied to first, second embodiment of information identifying method of film print on demand system based on the invention described above, step S20 specifically comprises:
Step S201, each digital template is read in circulation;
Step S202, contrasts the reference coordinate stored in the coordinate of the information identified region in film to be identified and digital template, determines the information identified region in film to be identified.
When identifying the identifying information in film to be identified, as the Wrong localization to information identified region, then in follow-up process, can not get correct identifying information, therefore correctly identified region location is the necessary condition obtaining correct identifying information.The subject matter identified in film information matching module in current film print on demand system is generally not identical for the electronic image memory device identified region of different layout, for certain image film any that different image documentation equipments produces, due at template construct working stage, to the identifying information region record all to some extent of its each layout, although can not determine which layout the image film of current identification belongs on earth, total energy ensure the coordinate of all records have at least one can be corresponding correct identified region.Therefore, when positioning the information identified region in film to be identified, the position configuration information in each digital template stored is read in circulation, the reference coordinate stored in the coordinate of the information identified region in film to be identified and digital template is contrasted, finally determines the information identified region in film to be identified.
With reference to the refinement schematic flow sheet that Fig. 4, Fig. 4 are step S30 in Fig. 1.
Be applied to first, second embodiment of information identifying method of film print on demand system based on the invention described above, step S30 specifically comprises:
Step S301, carries out binary conversion treatment to the image of film to be identified;
Step S302, determines the upper and lower border of the character in the information identified region of the film after binary conversion treatment and left and right border, reads single characters all in this information identified region.
There is the gray-scale value representing black white image brightness at each pixel place of gray level image, and its scope is 0 to 255, and under normal circumstances, the binary conversion treatment of gray level image can facilitate follow-up image processing work.Therefore in order to the identification of the segmentation and single character that enable follow-up film numerical character is more fast and effeciently carried out, electronic pastes sector-meeting gray level image need be converted into the binary image only containing 0 and 255 two kind of gray-scale value.
The bianry image obtained after binary conversion treatment, needs to carry out Character segmentation to isolate single character one by one, just can carry out the subsequent treatment such as feature extraction and Classification and Identification.Because the region, left and right of identification character and white space area grayscale value are 0, therefore can find to occur gray-scale value summation trough in the left and right blank space of character through vertical projection.By arranging suitable maximum trough threshold value, the gray-scale value summation that statistics image often arranges, thus the right boundary of character is come out.Concrete realization flow is: from left to right add up the vertical projection gray-scale value that often arranges of image and numerical value, in time running into first gray-scale value summation and be less than maximum trough threshold value, using the left margin of these row as Far Left character.Continue to carry out to the right, in time running into another row gray-scale value summation and be less than maximum trough threshold value, the right margin of these row as Far Left character.Method can be found out remaining character right boundary like this, but for the electronic image memory device image that different images equipment produces, the form of character and intercharacter gap are that difference is very large, determine that the gray-scale value that often arranges and minimum value and the appropriate character pitch threshold value that arranges will be the difficult points of Character segmentation.Another kind of situation occurs the phenomenon of Characters Stuck after the first Character segmentation of partial image, therefore needs to carry out second time segmentation and cut.
Concrete grammar is as follows:
1, character up-and-down boundary is determined:
1) from top to bottom (or from bottom to top) image is scanned in the horizontal direction, record the gray-scale value summation of every a line, the interference of electronic image memory device image identifying information noise region is very little.
2) when the gray-scale value summation of certain a line is more than or equal to some threshold values (being 4) herein, think and may have found the upper edge of numerical character, record line number now.
3) continue the pixel value summation of (upwards) statistics remaining row downwards, check whether the row also having and satisfy condition, until find the behavior do not satisfied condition to stop, namely this journey can think the lower limb of digital edge, is designated as this line number.
4) height of calculating character:
5) read the layout height information of preserving in template construct instrument, as >2, then think that the cutting of character upper-lower height makes mistakes.Return and read next topology data.
6) the area grayscale value that image line number is less than and line number is greater than all is set to 0.
2. character right boundary is determined:
1) image vertically scans by right from a left side (or left from the right side), and record the gray-scale value summation of each row, the interference of electronic image memory device image identifying information noise region is very little.
2) when the gray-scale value summation of a certain row is more than or equal to some threshold values (being 1) herein, think and may have found the left margin of numerical character, record columns now.
3) continue (left) to the right and add up the pixel value summation of all the other row, check whether the row also having and satisfy condition, until find the behavior do not satisfied condition to stop, namely this journey can think the right hand edge of digital edge, is designated as this line number.
4) width of calculating character:
5) read the layout width information of preserving in template construct instrument, as >2, then think to cut about character and make mistakes.Return and read next topology data.
6) continue to read, until all single characters mark off.Copy the determination of character height, the vertical method that character carries out from left to right is scanned, adds up the pixel value summation of each row.
When the width cutting the single character obtained is excessive, when being the twice of standard word fabric width degree substantially, think that this kind of situation is Characters Stuck.By improving row gray-scale value summation minimum value, segmentation being carried out to adhesion part and cuts.
With reference to the refinement schematic flow sheet that Fig. 5, Fig. 5 are step S40 in Fig. 1.
Be applied to first, second embodiment of information identifying method of film print on demand system based on the invention described above, step S40 specifically comprises:
Step S401, carries out character normalization process to the single character obtained after Character segmentation, and character normalization process comprises character position normalized and character boundary normalized;
Step S402, the reference character pixel in the pixel of the single character after contrast character normalization process and digital template, identifies whole single characters;
Step S403, the whole single character of linkage identification, obtains the identifying information in information identified region.
Due in the process that makes in digital template, all digital template sizes are unit picture element, but, due to different images equipment produce electronic image memory device in character size different, therefore the single character obtained in identifying needs to carry out the normalized of character, mainly comprises place normalization and size normalization.Due to the printed character that the identifying information in electronic image memory device image is all standard, therefore do not need to carry out place normalization to it, the present embodiment adopts the mode of the size of image proportionally being carried out the linear a certain size that zooms in or out to be normalized, the first scaling Radio of calculated level and vertical direction
x, Radio
y, computing method are as follows:
Wherein, the width before img_widthin and img_widthout is respectively character picture normalization and after normalization, the height before img_heightin and img_heightout is respectively character picture normalization and after normalization.
If certain point (x, y) after normalization on image g (x, y), corresponding to (u, the v) point in character picture f (x, y) former before normalization, before and after normalization, in image, the mapping equation of corresponding pixel points is:
The coordinate that pixel after such normalized on image is usually mapped to the point of original image is not integer, and this just needs to utilize interpolation algorithm to decide the value of this pixel.What adopt herein is bilinear interpolation, and the core of this algorithm carries out once linear interpolation respectively in vertical and horizontal direction.If the mapping point (μ, ν) obtained, between four pixels of original image, makes α=μ-[μ]; β=ν-[ν]; Wherein [] represents round numbers, carries out linear interpolation in the horizontal direction and obtains:
t
1＝αf([μ]+1,[ν])+(1-α)f([μ],[ν])；
t
2＝αf([μ]+1,[ν])+(1-α)f([μ],[ν]+1)；
Carry out numerical value directional interpolation again to after horizontal direction interpolation, obtain:
g(x,y)＝f(μ,ν)＝βt
2+(1-β)t
1
Utilize bilinear interpolation that single character image normalization is treated to uniform sizes.
Reference character pixel in the pixel of the single character after contrast character normalization process and digital template, identifies whole single characters.The present embodiment adopts the method for point-to-point coupling, is compared by the pixel value that single character to be identified and the template character of standard carry out corresponding point, and adds up final pixel difference value.Because the various film layouts of each image documentation equipment before have all made the digital template of standard, therefore carry out after point-to-point pixel compares to character and template, the minimum numeral of difference value is the identification character obtained.After each single character all identifies, coupling together, namely obtain basic discriminating digit result, on this basis by adding prefix and the suffix of this layout electronic image memory device in configuration information, namely obtaining final required identifying information result.
The present invention also provides a kind of information recognition device being applied to film print on demand system.
Reference Fig. 6, Fig. 6 are the high-level schematic functional block diagram that the present invention is applied to information recognition device first embodiment of film print on demand system.
In one embodiment, the information recognition device being applied to film print on demand system comprises:
Receiver module 10, for receiving the film to be identified that film print on demand system is uploaded;
Locating module 20, for positioning the information identified region in film to be identified according to the digital template preset;
Process and segmentation module 30, for carrying out image procossing to film to be identified, and carry out Character segmentation to the character in the information identified region of location;
Identify and determination module 40, for the single character obtained after identification character segmentation, according to the identifying information in the single character comformed information identified region identified.
The embodiment of the present invention can be applicable to various film print on demand system, and before identifying the identifying information in film to be identified, be provided with corresponding digital template for the issuable film of various different film print on demand systems and store, to be identified by the identifying information of digital template to the film to be identified received.After receiver module 10 receives the film to be identified that film print on demand system uploads, locating module 20 positions the information identified region in film to be identified according to the digital template preset, each digital template that concrete localization method stores by traversal, information identified region and digital template are compared, finally determines the position at the information identified region place in film to be identified.
Determine the information identified region in film to be identified, in order to the character in follow-up identifying information identified region, thus by processing and splitting module 30, image procossing is carried out to this film to be identified, and further Character segmentation is carried out to the character in the information identified region of location, then identify by identification and determination module 40 the single character obtained after Character segmentation one by one, and according to the identifying information in the final comformed information identified region of each single character identified, thus obtain the relevant identification information in pending film.
The present embodiment is after receiving the film to be identified that film print on demand system uploads, digital template according to presetting positions the information identified region in film to be identified, image procossing is carried out to film to be identified, and Character segmentation is carried out to the character in the information identified region of location; The single character obtained after identification character segmentation, according to the identifying information in the single character comformed information identified region identified.The feature of the film image information that embodiment of the present invention bonding apparatus workstation and diagnosis typesetting workstation produce, solve electronic image memory device that current film print on demand system produces for distinct device or the not good problem of the different layout film image recognition effects that same equipment produces, its applied range, high to the information recognition efficiency of film, and automatic, the accurate match of film and corresponding patient information can be realized.
Reference Fig. 7, Fig. 7 are the high-level schematic functional block diagram that the present invention is applied to information recognition device second embodiment of film print on demand system.
Based on above-described embodiment, propose second embodiment of the invention, this information recognition device being applied to film print on demand system also comprises:
Interface provides module 50, and for providing digital template to make interface, the identification parameter for the film to be identified generated according to each film print on demand system arranges digital template;
Preserve module 51, for preserving each digital template of setting.
Due to different image documentation equipments produce printing film identifying information residing for position different, for the electronic image memory device of the different printing layouts that same image documentation equipment produces, position residing for identifying information is also not identical usually, moreover, identification character morphological feature, identify length also by may be different.Therefore, before identification film, interface provides module 50 to provide digital template to make interface to user, the printing film produced for various different image documentation equipment by this digital template making interface for user makes digital template, while making digital template, record the position that in the film of each printing layout, identifying information is corresponding, namely all make set of number template and corresponding identification parameter for the printing film of different images equipment generation and the multiple layout film of same equipment.Each equipment workstation is sent in the configuration information of virtual printing service end, all comprise the information of an AETitle, this information unique identification sends the equipment workstation model of image, is determined the source device of this image before carrying out image identification by the AETitle item read in configuration information.
Before whole print on demand system drops into formal use, need to carry out the making of respective digital template and the record of configuration information to all layouts of each equipment of hospital come into operation, therefore before use system, need to carry out alternately with corresponding hospital, the film data of the various layouts that the image documentation equipment workstation commonly used to obtain them generates.Film data are normally transmitted with the form of DICOM file, and DICOM file is a kind of special file, not only contains image, and it also comprises the basic parameter of the electronic image memory device correction of image, as image width, highly, picture element density etc.After DICOM file is presented at digital template making interface, need to fill in the partial parameters information at interface, current partial image equipment, the information identified and patient ID or checking sequence number may be not only digital, but the combination of numeral and letter, the situation of letter and number cross arrangement seldom occurs, therefore, the feature little in conjunction with alphabetical probability of occurrence and equipment generation film image letter is identical, make in interface in digital template and prefix and suffix information are recorded, if identifying information is without prefix and suffix, correspondence position puts sky, identify the rear corresponding informance reading record, final recognition result is obtained by combination.
Simultaneously, digital template makes interface based on the message response function in MFC, and by amendment response function code, instrument is supported, and mouse frame favored area is converted to the function of coordinate, by recording the position coordinates of identified region after conversion, and save as follow-up use.After frame selects identified region to complete, the discriminating digit that input frame favored area is corresponding, and electronic image memory device that selection this time makes is which layout sequence of this equipment, only has current layout of choosing to be greater than just permission when there is layout.After the essential information of electronic image memory device necessity completes, generate Template button and complete the current digital template chosen by clicking and generate.Preserve module 51 when template construct completes to need to preserve the film configuration information of this layout, utilize the function defined in LiniFileConfig class to preserve configuration information.
Be applied to first, second embodiment of information recognition device of film print on demand system based on the invention described above, locating module 20 specifically for:
Each digital template is read in circulation;
The reference coordinate stored in the coordinate of the information identified region in film to be identified and digital template is contrasted, determines the information identified region in film to be identified.
When identifying the identifying information in film to be identified, as the Wrong localization to information identified region, then in follow-up process, can not get correct identifying information, therefore correctly identified region location is the necessary condition obtaining correct identifying information.The subject matter identified in film information matching module in current film print on demand system is generally not identical for the electronic image memory device identified region of different layout, for certain image film any that different image documentation equipments produces, due at template construct working stage, to the identifying information region record all to some extent of its each layout, although can not determine which layout the image film of current identification belongs on earth, total energy ensure the coordinate of all records have at least one can be corresponding correct identified region.Therefore, when positioning the information identified region in film to be identified, the position configuration information in each digital template stored is read in locating module 20 circulation, the reference coordinate stored in the coordinate of the information identified region in film to be identified and digital template is contrasted, finally determines the information identified region in film to be identified.
Be applied to first, second embodiment of information recognition device of film print on demand system based on the invention described above, process and segmentation module 30 specifically for:
Binary conversion treatment is carried out to the image of film to be identified;
Determine the upper and lower border of the character in the information identified region of the film after binary conversion treatment and left and right border, read single characters all in this information identified region.
There is the gray-scale value representing black white image brightness at each pixel place of gray level image, and its scope is 0 to 255, and under normal circumstances, the binary conversion treatment of gray level image can facilitate follow-up image processing work.Therefore in order to the identification of the segmentation and single character that enable follow-up film numerical character is more fast and effeciently carried out, electronic pastes sector-meeting gray level image need be converted into the binary image only containing 0 and 255 two kind of gray-scale value by process and segmentation module 30.
The bianry image obtained after binary conversion treatment, needs to carry out Character segmentation to isolate single character one by one, just can carry out the subsequent treatment such as feature extraction and Classification and Identification.Because the region, left and right of identification character and white space area grayscale value are 0, therefore can find to occur gray-scale value summation trough in the left and right blank space of character through vertical projection.By arranging suitable maximum trough threshold value, the gray-scale value summation that statistics image often arranges, thus the right boundary of character is come out.Concrete realization flow is: from left to right add up the vertical projection gray-scale value that often arranges of image and numerical value, in time running into first gray-scale value summation and be less than maximum trough threshold value, using the left margin of these row as Far Left character.Continue to carry out to the right, in time running into another row gray-scale value summation and be less than maximum trough threshold value, the right margin of these row as Far Left character.Method can be found out remaining character right boundary like this, but for the electronic image memory device image that different images equipment produces, the form of character and intercharacter gap are that difference is very large, determine that the gray-scale value that often arranges and minimum value and the appropriate character pitch threshold value that arranges will be the difficult points of Character segmentation.Another kind of situation occurs the phenomenon of Characters Stuck after the first Character segmentation of partial image, therefore needs to carry out second time segmentation and cut.
Concrete grammar is as follows:
1, character up-and-down boundary is determined:
1) from top to bottom (or from bottom to top) image is scanned in the horizontal direction, record the gray-scale value summation of every a line, the interference of electronic image memory device image identifying information noise region is very little.
2) when the gray-scale value summation of certain a line is more than or equal to some threshold values (being 4) herein, think and may have found the upper edge of numerical character, record line number now.
3) continue the pixel value summation of (upwards) statistics remaining row downwards, check whether the row also having and satisfy condition, until find the behavior do not satisfied condition to stop, namely this journey can think the lower limb of digital edge, is designated as this line number.
4) height of calculating character:
5) read the layout height information of preserving in template construct instrument, as >2, then think that the cutting of character upper-lower height makes mistakes.Return and read next topology data.
6) the area grayscale value that image line number is less than and line number is greater than all is set to 0.
2. character right boundary is determined:
1) image vertically scans by right from a left side (or left from the right side), and record the gray-scale value summation of each row, the interference of electronic image memory device image identifying information noise region is very little.
2) when the gray-scale value summation of a certain row is more than or equal to some threshold values (being 1) herein, think and may have found the left margin of numerical character, record columns now.
3) continue (left) to the right and add up the pixel value summation of all the other row, check whether the row also having and satisfy condition, until find the behavior do not satisfied condition to stop, namely this journey can think the right hand edge of digital edge, is designated as this line number.
4) width of calculating character:
5) read the layout width information of preserving in template construct instrument, as >2, then think to cut about character and make mistakes.Return and read next topology data.
6) continue to read, until all single characters mark off.Copy the determination of character height, the vertical method that character carries out from left to right is scanned, adds up the pixel value summation of each row.
When the width cutting the single character obtained is excessive, when being the twice of standard word fabric width degree substantially, think that this kind of situation is Characters Stuck.By improving row gray-scale value summation minimum value, segmentation being carried out to adhesion part and cuts.
Be applied to first, second embodiment of information recognition device of film print on demand system based on the invention described above, identify and determination module 40 specifically for:
Carry out character normalization process to the single character obtained after Character segmentation, character normalization process comprises character position normalized and character boundary normalized;
Reference character pixel in the pixel of the single character after contrast character normalization process and digital template, identifies whole single characters;
The whole single character of linkage identification, obtains the identifying information in information identified region.
Due in the process that makes in digital template, all digital template sizes are unit picture element, but, due to different images equipment produce electronic image memory device in character size different, therefore identification and determination module 40 need the normalized of the single character obtained in identification being carried out to character, mainly comprise place normalization and size normalization.Due to the printed character that the identifying information in electronic image memory device image is all standard, therefore do not need to carry out place normalization to it, the present embodiment adopts the mode of the size of image proportionally being carried out the linear a certain size that zooms in or out to be normalized, the first scaling Radio of calculated level and vertical direction
x, Radio
y, computing method are as follows:
Wherein, the width before img_widthin and img_widthout is respectively character picture normalization and after normalization, the height before img_heightin and img_heightout is respectively character picture normalization and after normalization.
If certain point (x, y) after normalization on image g (x, y), corresponding to (u, the v) point in character picture f (x, y) former before normalization, before and after normalization, in image, the mapping equation of corresponding pixel points is:
The coordinate that pixel after such normalized on image is usually mapped to the point of original image is not integer, and this just needs to utilize interpolation algorithm to decide the value of this pixel.What adopt herein is bilinear interpolation, and the core of this algorithm carries out once linear interpolation respectively in vertical and horizontal direction.If the mapping point (μ, ν) obtained, between four pixels of original image, makes α=μ-[μ]; β=ν-[ν]; Wherein [] represents round numbers, carries out linear interpolation in the horizontal direction and obtains:
t
1＝αf([μ]+1,[ν])+(1-α)f([μ],[ν])；
t
2＝αf([μ]+1,[ν])+(1-α)f([μ],[ν]+1)；
Carry out numerical value directional interpolation again to after horizontal direction interpolation, obtain:
g(x,y)＝f(μ,ν)＝βt
2+(1-β)t
1
Utilize bilinear interpolation that single character image normalization is treated to uniform sizes.
Reference character pixel in the pixel of the single character after contrast character normalization process and digital template, identifies whole single characters.The present embodiment adopts the method for point-to-point coupling, is compared by the pixel value that single character to be identified and the template character of standard carry out corresponding point, and adds up final pixel difference value.Because the various film layouts of each image documentation equipment before have all made the digital template of standard, therefore carry out after point-to-point pixel compares to character and template, the minimum numeral of difference value is the identification character obtained.After each single character all identifies, coupling together, namely obtain basic discriminating digit result, on this basis by adding prefix and the suffix of this layout electronic image memory device in configuration information, namely obtaining final required identifying information result.
These are only the preferred embodiments of the present invention; not thereby the scope of the claims of the present invention is limited; every utilize instructions of the present invention and accompanying drawing content to do equivalent structure or equivalent flow process conversion; or be directly or indirectly used in other relevant technical fields, be all in like manner included in scope of patent protection of the present invention.
Claims (10)
1. be applied to an information identifying method for film print on demand system, it is characterized in that, described in be applied to film print on demand system information identifying method comprise the steps:
Receive the film to be identified that film print on demand system is uploaded;
Digital template according to presetting positions the information identified region in described film to be identified;
Image procossing is carried out to described film to be identified, and Character segmentation is carried out to the character in the described information identified region of location;
The single character obtained after identification character segmentation, determines the identifying information in described information identified region according to the described single character identified.
2. be applied to the information identifying method of film print on demand system as claimed in claim 1, it is characterized in that, before the step of the film to be identified that described reception film print on demand system is uploaded, described in be applied to film print on demand system information identifying method also comprise:
There is provided digital template to make interface, the identification parameter for the film to be identified generated according to each film print on demand system arranges digital template;
Digital template described in each of preservation setting.
3. be applied to the information identifying method of film print on demand system as claimed in claim 1 or 2, it is characterized in that, the digital template that described basis is preset comprises the step that the information identified region in described film to be identified positions:
Digital template described in each is read in circulation;
The reference coordinate stored in the coordinate of the information identified region in described film to be identified and described digital template is contrasted, determines the information identified region in film to be identified.
4. be applied to the information identifying method of film print on demand system as claimed in claim 1 or 2, it is characterized in that, described image procossing is carried out to film to be identified, and the step that the character in the described information identified region of location carries out Character segmentation is comprised:
Binary conversion treatment is carried out to the image of described film to be identified;
Determine the upper and lower border of the character in the information identified region of the described film after binary conversion treatment and left and right border, read single characters all in this information identified region.
5. be applied to the information identifying method of film print on demand system as claimed in claim 1 or 2, it is characterized in that, the single character obtained after the segmentation of described identification character, the step of the identifying information determined in described information identified region according to the described single character identified comprises:
Carry out character normalization process to the single character obtained after Character segmentation, described character normalization process comprises character position normalized and character boundary normalized;
Reference character pixel in the pixel of the described single character after contrast character normalization process and described digital template, identifies whole single characters;
The whole described single character of linkage identification, obtains the identifying information in described information identified region.
6. be applied to an information recognition device for film print on demand system, it is characterized in that, described in be applied to film print on demand system information recognition device comprise:
Receiver module, for receiving the film to be identified that film print on demand system is uploaded;
Locating module, for positioning the information identified region in described film to be identified according to the digital template preset;
Process and segmentation module, for carrying out image procossing to described film to be identified, and carry out Character segmentation to the character in the described information identified region of location;
Identify and determination module, for the single character obtained after identification character segmentation, determine the identifying information in described information identified region according to the described single character identified.
7. be applied to the information recognition device of film print on demand system as claimed in claim 6, it is characterized in that, described in be applied to film print on demand system information recognition device also comprise:
Interface provides module, and for providing digital template to make interface, the identification parameter for the film to be identified generated according to each film print on demand system arranges digital template;
Preserve module, for preserve setting each described in digital template.
8. the information recognition device being applied to film print on demand system as claimed in claims 6 or 7, is characterized in that, described locating module specifically for:
Digital template described in each is read in circulation;
The reference coordinate stored in the coordinate of the information identified region in described film to be identified and described digital template is contrasted, determines the information identified region in film to be identified.
9. the information recognition device being applied to film print on demand system as claimed in claims 6 or 7, is characterized in that, described process and segmentation module specifically for:
Binary conversion treatment is carried out to the image of described film to be identified;
Determine the upper and lower border of the character in the information identified region of the described film after binary conversion treatment and left and right border, read single characters all in this information identified region.
10. the information recognition device being applied to film print on demand system as claimed in claims 6 or 7, is characterized in that, described identification and determination module specifically for:
Carry out character normalization process to the single character obtained after Character segmentation, described character normalization process comprises character position normalized and character boundary normalized;
Reference character pixel in the pixel of the described single character after contrast character normalization process and described digital template, identifies whole single characters;
The whole described single character of linkage identification, obtains the identifying information in described information identified region.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510491352.6A CN105160343A (en) | 2015-08-12 | 2015-08-12 | Information identification method and device applied to film on-demand-printing system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510491352.6A CN105160343A (en) | 2015-08-12 | 2015-08-12 | Information identification method and device applied to film on-demand-printing system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105160343A true CN105160343A (en) | 2015-12-16 |
Family
ID=54801195
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510491352.6A Pending CN105160343A (en) | 2015-08-12 | 2015-08-12 | Information identification method and device applied to film on-demand-printing system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105160343A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106874681A (en) * | 2017-02-28 | 2017-06-20 | 宋爱华 | One kind is used for surgery image data analytic method, system and full institute's Self-service printing system |
| CN108734167A (en) * | 2018-05-08 | 2018-11-02 | 湖南开启时代电子信息技术有限公司 | A kind of contaminated film character recognition method |
| CN109145879A (en) * | 2018-09-30 | 2019-01-04 | 金蝶软件（中国）有限公司 | A kind of type fount knows method for distinguishing, equipment and storage medium |
| CN110060270A (en) * | 2019-04-25 | 2019-07-26 | 宁锐慧创信息科技南京有限公司 | A kind of edge detection method suitable for low image quality multiangular tubular object |
| CN111292837A (en) * | 2020-01-13 | 2020-06-16 | 上海杏脉信息科技有限公司 | Medical image auxiliary diagnosis system and triggering method, device, medium and equipment thereof |
| CN111613305A (en) * | 2020-05-27 | 2020-09-01 | 周兴祥 | Method and system for generating DICOMDIR file |
| CN112233775A (en) * | 2020-09-30 | 2021-01-15 | 宁波市科技园区明天医网科技有限公司 | Method for automatically generating check matching sample film for electronic film |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101800769A (en) * | 2010-01-19 | 2010-08-11 | 北京新网医讯技术有限公司 | Electric film printing system and method based on DICOM |
| CN101913297A (en) * | 2010-06-25 | 2010-12-15 | 杭州电子科技大学 | Medical film print on demand control method |
| CN201728914U (en) * | 2010-06-25 | 2011-02-02 | 杭州电子科技大学 | Medical film on-demand printing control device |
| CN103332013A (en) * | 2013-06-17 | 2013-10-02 | 广东柯丽尔新材料有限公司 | System and method for DICOM-based on-demand printing |
-
2015
- 2015-08-12 CN CN201510491352.6A patent/CN105160343A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101800769A (en) * | 2010-01-19 | 2010-08-11 | 北京新网医讯技术有限公司 | Electric film printing system and method based on DICOM |
| CN101913297A (en) * | 2010-06-25 | 2010-12-15 | 杭州电子科技大学 | Medical film print on demand control method |
| CN201728914U (en) * | 2010-06-25 | 2011-02-02 | 杭州电子科技大学 | Medical film on-demand printing control device |
| CN103332013A (en) * | 2013-06-17 | 2013-10-02 | 广东柯丽尔新材料有限公司 | System and method for DICOM-based on-demand printing |
Non-Patent Citations (3)
| Title |
|---|
| 徐世才: "胶片按需打印技术研究及***实现", 《中国优秀硕士学位论文全文数据库 医药卫生科技辑》 * |
| 徐彦栋: "基于DICOM标准的按需打印***设计和实现", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 王伯胤 等: "胶片按需打印、自助式取片在优化放射检查工作流程的应用", 《中国医院》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106874681A (en) * | 2017-02-28 | 2017-06-20 | 宋爱华 | One kind is used for surgery image data analytic method, system and full institute's Self-service printing system |
| CN108734167A (en) * | 2018-05-08 | 2018-11-02 | 湖南开启时代电子信息技术有限公司 | A kind of contaminated film character recognition method |
| CN109145879A (en) * | 2018-09-30 | 2019-01-04 | 金蝶软件（中国）有限公司 | A kind of type fount knows method for distinguishing, equipment and storage medium |
| CN109145879B (en) * | 2018-09-30 | 2021-01-12 | 金蝶软件（中国）有限公司 | Method, equipment and storage medium for identifying printing font |
| CN110060270A (en) * | 2019-04-25 | 2019-07-26 | 宁锐慧创信息科技南京有限公司 | A kind of edge detection method suitable for low image quality multiangular tubular object |
| CN110060270B (en) * | 2019-04-25 | 2021-05-04 | 宁锐慧创信息科技南京有限公司 | Edge detection method suitable for polygonal tubular object with low imaging quality |
| CN111292837A (en) * | 2020-01-13 | 2020-06-16 | 上海杏脉信息科技有限公司 | Medical image auxiliary diagnosis system and triggering method, device, medium and equipment thereof |
| CN111613305A (en) * | 2020-05-27 | 2020-09-01 | 周兴祥 | Method and system for generating DICOMDIR file |
| CN111613305B (en) * | 2020-05-27 | 2024-04-30 | 周兴祥 | DICOMDIR file generation method and system |
| CN112233775A (en) * | 2020-09-30 | 2021-01-15 | 宁波市科技园区明天医网科技有限公司 | Method for automatically generating check matching sample film for electronic film |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105160343A (en) | Information identification method and device applied to film on-demand-printing system | |
| US20210256253A1 (en) | Method and apparatus of image-to-document conversion based on ocr, device, and readable storage medium | |
| CN109522816B (en) | Table identification method and device and computer storage medium | |
| CN102567300B (en) | Picture document processing method and device | |
| CN110069767B (en) | Typesetting method based on electronic book, electronic equipment and computer storage medium | |
| CN110443140B (en) | Text positioning method, device, computer equipment and storage medium | |
| CN104008384A (en) | Character identification method and character identification apparatus | |
| CN105389165A (en) | Document picture display method and apparatus, and terminal | |
| CN112446262A (en) | Text analysis method, text analysis device, text analysis terminal and computer-readable storage medium | |
| CN107016417A (en) | A kind of method and device of character recognition | |
| CN108319578B (en) | Method for generating medium for data recording | |
| CN112712014A (en) | Table picture structure analysis method, system, equipment and readable storage medium | |
| CN111104883A (en) | Job answer extraction method, device, equipment and computer readable storage medium | |
| CN110705442B (en) | Automatic test paper answer acquisition method, terminal equipment and storage medium | |
| CN112926421A (en) | Image processing method and apparatus, electronic device, and storage medium | |
| JP4626777B2 (en) | Information processing apparatus and information processing program | |
| CN111428446B (en) | Questionnaire identification method and questionnaire identification system | |
| CN113537184A (en) | OCR (optical character recognition) model training method and device, computer equipment and storage medium | |
| CN108804978B (en) | Layout analysis method and device | |
| CN113591827A (en) | Text image processing method and device, electronic equipment and readable storage medium | |
| CN112183294A (en) | Text elimination method and device, electronic equipment and storage medium | |
| US20140320527A1 (en) | Hardware glyph cache | |
| CN115797939A (en) | Two-stage italic character recognition method and device based on deep learning | |
| CN112084103A (en) | Interface test method, device, equipment and medium | |
| CN113936187A (en) | Text image synthesis method and device, storage medium and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20151216 |
|
| RJ01 | Rejection of invention patent application after publication |