CN106547760B - Image retrieval system for leather cloth color card - Google Patents
Image retrieval system for leather cloth color card Download PDFInfo
- Publication number
- CN106547760B CN106547760B CN201510598398.8A CN201510598398A CN106547760B CN 106547760 B CN106547760 B CN 106547760B CN 201510598398 A CN201510598398 A CN 201510598398A CN 106547760 B CN106547760 B CN 106547760B
- Authority
- CN
- China
- Prior art keywords
- image
- color card
- retrieval
- executing
- module
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
- G06F16/5838—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using colour
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
Landscapes
- Engineering & Computer Science (AREA)
- Library & Information Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Processing Or Creating Images (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention provides a leather cloth color card image retrieval system, which comprises: an image acquisition module: the system is used for acquiring an image to be retrieved and formulating a feature extraction plan and a matching rule according to selection combination of visual features and selection of a matching precision threshold; an image feature extraction module: the visual characteristic of the image to be retrieved is extracted and stored in an image characteristic library; an image feature granularity calculation module: compatible granularity used for dividing the image; the image retrieval matching module: the image matching system is used for calculating the distance between the image to be retrieved and each image feature in the image feature library and screening the images meeting the matching precision threshold; the color card information modification module: the system is used for modifying and storing the color card information in the color card library; the retrieval result display module: for displaying the search results. The leather cloth color card image retrieval system has the following beneficial effects: the workload is small, the consumed time is short, the error is reduced, and convenience is brought to leather and cloth suppliers.
Description
Technical Field
The invention relates to the field of leather cloth retrieval, in particular to a color card image retrieval system for leather cloth.
Background
The bag factories need to purchase raw materials frequently and in large quantities, and numerous leather cloth suppliers need to search various leather cloths meeting the needs of customers from a large number of leather cloth color cards. The case and bag factory often needs to look for the material, the present situation is that people are dispatched everyday to run the market to look for, so there is no target, its work load is great, and it is long to spend time, for the supplier's grade, because there are many employees as new employees, and not necessarily familiar with all the materials sold, can only rely on personal feeling, when a buyer comes to grade to ask there is no necessary material, the business of the supplier only sees and answers at once, there is or not, so it is easy to lose business, meanwhile, because the supplier needs to look for the leather cloth manually, it causes the mistake easily, bring many inconveniences to the leather cloth supplier at the same time. Therefore, bag shop purchasers and leather clothing suppliers urgently need a computer application system for conveniently and effectively retrieving the leather clothing color card image.
Disclosure of Invention
The technical problem to be solved by the invention is to provide a leather cloth color card image retrieval system which has the advantages of small workload, short consumed time, error reduction and convenience for leather cloth suppliers, aiming at the defects of large workload, long consumed time and inconvenience for leather cloth suppliers caused by errors in the prior art.
The technical scheme adopted by the invention for solving the technical problems is as follows: constructing a leather cloth color card image retrieval system comprising:
an image acquisition module: the system comprises a database, a matching precision threshold value and a characteristic extraction plan, wherein the database is used for storing a plurality of visual characteristics of a user;
an image feature extraction module: the system is used for extracting the visual features of the image to be retrieved by combining the selected visual feature extraction method according to the selection combination of the user on the visual features and the selection of the matching precision threshold value, and storing the extracted visual feature multidimensional vector into an image feature library;
an image feature granularity calculation module: the compatibility granularity used for dividing the image by the domain knowledge input by the user and the extracted visual features;
the image retrieval matching module: the image matching method is used for calculating the distance between the image to be retrieved and the characteristics of each image in the image characteristic library according to the selected similarity calculation method, and screening out the images which accord with the matching precision threshold;
the color card information modification module: the system is used for modifying and storing the color card information in the color card library and verifying and detecting the modified value;
the retrieval result display module: and the system is used for displaying the retrieval result and carrying out secondary screening according to the screening condition of the user.
In the leather cloth color card image retrieval system of the present invention, further comprising:
a system initialization module: the system is used for initializing all parameters of the leather cloth color card image retrieval system;
a system setting module: modifying various parameters of the leather cloth color card image retrieval system;
an image interception module: the image acquisition device is used for intercepting the size of an image in the image acquisition process;
color card deletion module: the system is used for deleting the color card images individually and in batches;
a character retrieval module: the method is used for searching the images in the color card library through the keyword information.
In the leather cloth color card image retrieval system of the present invention, the image acquisition module, when acquiring an image, comprises the steps of:
A) judging whether system import collection or scanning collection is adopted, and if scanning collection is adopted, executing the step B); if the system import collection is adopted, executing the step C);
B) setting scanning parameters and executing the step C); the scanning parameters comprise size, resolution and storage format;
C) displaying the acquired image;
D) judging whether the acquired image is subjected to automatic screenshot or manual screenshot, and if the acquired image is subjected to the manual screenshot, executing the step E); if the screenshot is automatically captured, executing step F);
E) setting scanning parameters again, and executing the step F); the scanning parameters comprise size, resolution and storage format;
F) displaying the screenshot image, and judging whether the screenshot image meets the requirements, if so, executing the step G); otherwise, returning to the step D);
G) judging whether to perform image warehousing or image scanning comparison, and if so, executing the step H); if the image scanning comparison is carried out, executing the step I);
H) inputting image information, and executing the step I); the image information comprises a color card name, a color card number and a supplier;
I) and warehousing the image, and displaying the image after screenshot.
In the leather cloth color card image retrieval system of the present invention, the image feature extraction module, when performing image feature extraction, includes the following steps:
A1) carrying out image preprocessing; the image preprocessing comprises gray level histogram equalization, image smoothing and normalization operations;
B1) selecting a feature extraction option and judging whether the texture features are considered, if so, executing the step C1); otherwise, performing step D1);
C1) extracting image texture features, and executing step D1);
D1) extracting image color features;
E1) and normalizing the image features and storing the extracted features into the image feature library.
In the leather cloth color card image retrieval system of the present invention, the image retrieval matching module comprises the following steps when performing image retrieval matching:
A2) judging whether the simple retrieval or the advanced retrieval is selected, if so, executing the step B2); if so, perform step C2);
B2) setting a retrieval condition, and executing the step C2); the retrieval conditions comprise image types, retrieval numbers and retrieval precision;
C2) extracting the characteristics of the image to be matched;
D2) performing similarity measurement on the image to be matched and each image in the image feature library, and judging whether secondary screening operation is performed, if so, executing step E2); otherwise, performing step F2);
E2) setting secondary screening conditions, and executing step F2); the secondary screening conditions comprise leather cloth material, supplier and serial number;
F2) and displaying the retrieval result and finishing the retrieval matching operation.
In the leather cloth color card image retrieval system of the present invention, the retrieval result display module, when displaying the retrieval result, comprises the following steps:
A3) initializing a display panel;
B3) judging whether to select the search result display or the management display, if the search result display is performed, executing step C3); if yes, executing step D3);
C3) acquiring retrieval screening conditions and executing step D3);
D3) displaying the result, and finishing displaying the color card image and the related material information; the related material information comprises supplier source, price and contact information.
In the leather cloth color card image retrieval system of the present invention, the color card information modification module, when modifying the color card information, comprises the steps of:
A4) opening a color card information modification page, and judging whether to modify the color card which is put into storage or not, if so, executing the step B4); if the warehoused color card is selected to be modified, executing the step C4);
B4) acquiring the existing information of the color cards in the color card library, and executing the step C4);
C4) filling or modifying the color card information and verifying the validity of the filled or modified color card information, if so, executing step D4); otherwise, filling or modifying the color card information again; the color card information comprises supplier source, price and contact information;
D4) and storing the filled or modified color card information into the color card library.
In the leather cloth color card image retrieval system of the present invention, the system initialization module, when performing system initialization, includes the steps of:
A5) detecting a system configuration file;
B5) judging whether the installation is the first installation, if so, executing the step C5); otherwise, performing step F5);
C5) installing a database;
D5) configuring a local file;
E5) initializing a system interface;
F5) and displaying a system interface.
In the leather cloth color card image retrieval system of the present invention, the system setting module, when performing system setting, includes the steps of:
A6) displaying a setting panel, and judging whether to restore the default setting, if so, executing the step D6); otherwise, performing step B6);
B6) inputting setting information, and executing step C6);
C6) judging whether the input setting information is valid, if so, executing step D6); otherwise, returning to the step B6);
D6) and saving the setting information.
The leather cloth color card image retrieval system has the following beneficial effects: the image retrieval method comprises the steps of obtaining an image to be retrieved by using an image acquisition module, an image feature extraction module, an image feature granularity calculation module, an image retrieval matching module, a color card information modification module and a retrieval result display module, wherein the image acquisition module is used for obtaining the image to be retrieved; the color card information in the color card library can be modified by using the color card information modification module; the purchasing personnel only need to take the leather cloth segment color card to each raw material supplier on the supply market where the leather cloth color card image retrieval system is installed, and the similar leather cloth color card is collected and retrieved through the leather cloth color card image retrieval system, so that the searching work of the leather cloth color card is simple and convenient, the workload is small, the time consumption is short, the error is reduced, and the convenience is brought to the leather cloth supplier.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a schematic view of an embodiment of a leather cloth color card image retrieval system according to the present invention;
FIG. 2 is a flow chart of the image acquisition module in the embodiment during image acquisition;
FIG. 3 is a flowchart of the image feature extraction module during image feature extraction in the embodiment;
FIG. 4 is a flowchart of the image retrieval matching module in the embodiment when performing image retrieval matching;
FIG. 5 is a flowchart of the search result display module in the embodiment when displaying the search result;
FIG. 6 is a flow chart of the color chart information modification module in modifying color chart information in the embodiment;
FIG. 7 is a flowchart of the system initialization module during system initialization according to the embodiment;
fig. 8 is a flowchart of the system setting module in the embodiment when performing system setting.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In the embodiment of the present invention, a schematic structural diagram of the leather cloth color card image retrieval system is shown in fig. 1. In fig. 1, the leather cloth color card image retrieval system includes an image acquisition module 1, an image feature extraction module 2, an image feature granularity calculation module 3, an image retrieval matching module 4, a color card information modification module 5 and a retrieval result display module 6; the image acquisition module 1 is used for acquiring an image to be retrieved input by a user, and making a next-step feature extraction plan and a matching rule according to a selection combination of the user on visual features and selection of a matching precision threshold; meanwhile, the granularity can be adjusted, so that different and more suitable retrieval operations can be performed according to different image categories. The image feature extraction module 2 is used for extracting the visual features of the image to be retrieved by combining the selected visual feature extraction method according to the selection combination of the user on the visual features and the selection of the matching precision threshold in the image acquisition module 1, and storing the extracted visual feature multidimensional vector into an image feature library; meanwhile, the image feature extraction work in the warehousing process is also within the working range of the image feature extraction module 2.
The image feature granularity calculation module 3 is used for dividing the compatible granularity of the image according to the domain knowledge input by the user and the extracted visual features; specifically, the present embodiment performs classification of multiple granularities on a leather cloth image by using a concept of particle computation, and performs classification by combining domain knowledge of the leather cloth image and color features, texture features, and the like of the leather cloth image. The efficiency is improved by seeking approximate solutions at different granularities, and the robustness is stronger because different granularities are adopted.
The image retrieval matching module 4 is used for calculating the distance between the image to be retrieved and the features of each image in the image feature library according to the selected similarity calculation method, and screening out the images which meet the matching precision threshold. The color card information modification module 5 is used for modifying and storing the color card information in the color card library and verifying and detecting the modified value. And the retrieval result display module 6 is used for displaying the retrieval result and carrying out secondary screening according to the screening conditions of the user. The purchasing personnel only need to take the leather cloth segment color card to each raw material supplier on the supply market where the leather cloth color card image retrieval system is installed, and the similar leather cloth color card is collected and retrieved through the leather cloth color card image retrieval system, so that the searching work of the leather cloth color card is simple and convenient, the workload is small, the time consumption is short, the error is reduced, and the convenience is brought to the leather cloth supplier.
In this embodiment, the leather cloth color card image retrieval system further includes a system initialization module 7, a system setting module 8, an image interception module 9, a color card deletion module 10, and a text retrieval module 11; the system initialization module 7 is used for initializing various parameters of the leather cloth color card image retrieval system; the system setting module 8 is used for modifying various parameters of the leather cloth color card image retrieval system; the image intercepting module 9 is used for intercepting the size of an image in the image acquisition process; the color chip deleting module 10 is used for deleting the color chip images individually and in batches; the character retrieval module 11 is used for retrieving the images in the color card library through the keyword information.
Fig. 2 is a flowchart of the image capturing module in this embodiment during image capturing. In fig. 2, the image capturing module 1 includes the following steps when capturing an image:
step S001, judging whether system import collection or scanning collection is adopted: in this step, it is determined whether to adopt system import acquisition or scan acquisition, and if scan acquisition is adopted, step S002 is executed; if the system import collection is adopted, step S003 is executed.
It should be noted that in this embodiment, the scanner is used to collect the image, and in order to enable different scanners to communicate with the computer, a uniform communication protocol, that is, a twain (toolkit withou An intervening name) protocol, is specified between the software developer and the hardware manufacturer. Most scanner manufacturers now support this communication protocol. In this embodiment, the interface provided by the TWAIN protocol is used to add the image acquisition module 1 to the leather cloth color card image retrieval system, so that the intermediate step of retrieving after importing the image acquired from the scanner software into the image retrieval system is avoided, and the scanning parameters are directly set in the leather cloth color card image retrieval system for scanning and then directly performing warehousing comparison. Software supporting the TWAIN protocol mainly contains three elements: application software, a data source manager and a data source. The TWAIN protocol takes a data source manager as a center, provides development interfaces for an application program and a data source respectively, and defines that image input is realized in a session process. The application completes the session process through an ENTRY function DSM _ ENTRY, an Operation Triplet (Operation Triplet) and a message response function of the TWAIN message and application provided by the data source manager.
Step S002 sets scanning parameters: in this step, scanning parameters are set, and the scanning parameters include the size, resolution, storage format, and the like of the scanned image. After the present step is performed, step S003 is performed.
Step S003 displays the acquired image: in this step, the acquired image is displayed. It is worth mentioning that the user can perform screenshot operation on the obtained image, and capture the part with obvious texture feature or color feature in the image to be matched for storage and comparison.
Step S004 determines whether to perform automatic screenshot or manual screenshot on the acquired image: in this step, it is determined whether to perform automatic screenshot or manual screenshot on the acquired image, and if the screenshot is a manual screenshot, step S005 is performed; if the screen capture is automatic, step S006 is performed.
Step S005 sets the scanning parameters again: in this step, the scan parameters including the scan image size, resolution, and save format, etc. are set again. After the present step is executed, step S006 is executed.
Step S006 displays the captured image, and judges whether the image meets the requirement: in this step, the image after screenshot is displayed, and whether the image meets the requirement is judged, if so, step S007 is executed; otherwise, return to step S004.
Step S007 judges whether to perform image warehousing or image scanning comparison: if the judgment result of the above step S006 is yes, the present step is executed. In this step, it is judged whether to perform image warehousing or image scanning comparison, and if so, step S008 is executed; if the comparison is image scanning comparison, step S010 is executed.
Step S008 input image information: if the judgment result of the above step S007 is yes, the present step is executed. In this step, image information is input, and the image information includes a color chart name, a color chart number, and a supplier. After the present step is executed, step S009 is executed.
Step S009 of putting the image into a warehouse and displaying the image after screenshot: in this step, the image is put in storage, and the image after screenshot is displayed.
Fig. 3 is a flowchart of the image feature extraction module in this embodiment when performing image feature extraction, and in fig. 3, the image feature extraction module 2 when performing image feature extraction includes the following steps:
step S101 performs image preprocessing: in this step, image preprocessing is performed, which includes gray histogram equalization, image smoothing, and normalization operations. The histogram equalization algorithm adopted by the embodiment is one of the most common and important algorithms in the image enhancement spatial domain algorithm. The method takes probability theory as a basis, and realizes the transformation of the histogram by using gray point operation, thereby achieving the purpose of image enhancement.
The embodiment adopts a TV (total variation) denoising method to perform denoising operation on the acquired image. The TV method is a popular image denoising method with a good denoising effect at present. The method achieves the purpose of smooth denoising by minimizing an energy function.
Step S102, selecting a feature extraction option, and judging whether texture features are considered: in this step, the selection of the feature extraction option is performed, and whether the texture feature is considered is judged, if the judgment result is yes, step S103 is executed; otherwise, step S104 is executed.
Step S103, extracting image texture features, wherein if the judgment result of the step S102 is positive, the step is executed, in the step, the image texture features are extracted, because the leather cloth belongs to artificial textures, the embodiment selects a K. L aws energy measurement method in a structure theory image texture representation method which is more beneficial to artificial texture retrieval, a method for gray level co-occurrence moment which has better retrieval capability for both natural and artificial textures, and an image texture representation method of wavelet transformation to extract the texture features of the image, and the step S104 is executed after the step is executed.
Step S104, extracting image color features: in this step, image color features are extracted. In the embodiment, an improved color space quantization method based on a dominant color clustering method is adopted, the attribution of colors is determined by calculating the distance, and meanwhile, 72-level image gray level degradation operation is performed by utilizing the characteristics of the HSV color space. In addition, as for leather cloth images, the colors generally include several to more than ten colors, and the types of the colors are not many, the present embodiment describes the images by using the color features based on the dominant color weight histogram.
Step S105 normalizes the image features, and stores the extracted features in an image feature library: in this step, the image features are normalized, and the extracted features are stored in an image feature library, that is, the image texture features and the color features are normalized, so that the obtained color card images with different sizes and different resolutions are normalized, and the extracted features are stored in the image feature library.
Fig. 4 is a flowchart of the image retrieval matching module in the present embodiment when performing image retrieval matching, and in fig. 4, the image retrieval matching module 4 includes the following steps when performing image retrieval matching:
step S201 determines whether to select simple retrieval or advanced retrieval: in this step, it is determined whether to select a simple search or a high-level search, and if the high-level search is performed, step S202 is executed; if so, step S203 is executed.
Step S202 sets the search condition: in this step, search conditions are set, and the search conditions include image type, search number, and search accuracy.
Step S203 extracts features of the image to be matched: in this step, the features of the image to be matched (image to be retrieved) are extracted.
Step S204, the similarity measurement is carried out on the image to be matched and each image in the image feature library, and whether the secondary screening operation is carried out or not is judged: in this step, the similarity measurement is performed on the image to be matched and each image in the image feature library, whether secondary screening operation is performed is judged, and if the judgment result is yes, step S205 is executed; otherwise, step S206 is executed. It should be noted that the similarity measure method adopted in the present embodiment is a measure combining the concept of consistent particles in particle calculation and the distance between feature components. And obtaining the classification information of the image to be retrieved through extracting the domain knowledge of the image to be retrieved and the color and texture characteristics of the image. Firstly, through division of domain knowledge, obtaining an image which is the same as the material (domain knowledge such as processing technology) of the leather cloth image to be retrieved, and obtaining a compatible image in a library; then, through the retrieval operation of the color granularity, the image in the library compatible with the color granularity is obtained, and the process comprises the granularity operation of the color dominant color number and the operation of the color containing relation granularity; if the retrieval result is too much and too complicated, further division is carried out by combining the granularity of the image texture features to obtain the compatible granularity combined with the image texture features, and finally the most similar matching result is obtained. If the minimum granularity is adopted for division, and the obtained retrieval result is empty, the retrieval condition is more relaxed by reducing the granularity matching condition, and a better retrieval result is obtained. In the process, the texture feature component of the image is subjected to matching screening operation by combining with distance operation.
Step S205 sets secondary screening conditions: if the judgment result of the step S204 is yes, the step is executed. In this step, secondary screening conditions (i.e., preset some screening conditions) are set, and the secondary screening conditions include leather cloth material, suppliers, and numbers.
Step S206 displays the search result, and ends the search matching operation: in this step, the search result is displayed, and the search matching operation is ended.
Fig. 5 is a flowchart of the search result display module in this embodiment when displaying the search result, and in fig. 5, the search result display module 6 includes the following steps when displaying the search result:
step S301 initializes the display panel: in this step, the display panel is initialized.
Step S302 judges whether to select search result display or management display: in this step, it is determined whether to select the search result display or the management display, and if the search result display is performed, step S303 is executed; if the management display is yes, step S304 is executed.
Step S303 acquires search screening conditions: in this step, the search screening conditions are obtained. After this step is executed, step S304 is executed.
S304, displaying the result, and finishing displaying the color card image and the related material information: in this step, the result is displayed, and the display of the color card image and the related material information is finished, wherein the related material information comprises supplier source, price and contact way.
Fig. 6 is a flowchart of the color chart information modifying module in this embodiment when modifying color chart information, and in fig. 6, the color chart information modifying module 5 includes the following steps when modifying color chart information:
step S401 is to open a color card information modification page, and to determine whether to select to modify the color card which is put in storage or not to be put in storage: in this step, a color card information modification page is opened, and whether the color card which is put into storage is selected to be modified or the color card which is not put into storage is selected to be modified is judged, if the color card which is not put into storage is selected to be modified, the step S402 is executed; if the color card in the warehouse is selected to be modified, step S403 is executed.
Step S402, obtaining the existing information of the color cards in the color card library: in this step, the existing information of the color cards in the color card library is obtained, and step S403 is executed after this step is executed.
Step S403 fills or modifies the color chip information, and verifies the validity of the filled or modified color chip information: in this step, the color card information is filled or modified, and the validity of the filled or modified color card information is verified, if valid, step S404 is executed; otherwise, the filling or the modification of the color card information in the step is carried out again. The color card information includes supplier source, price and contact information.
Step S404 stores the filled or modified color chart information, and stores it in the color chart library: in this step, the filled or modified color chip information is stored and stored in the color chip library.
Fig. 7 is a flowchart of the system initialization module in this embodiment when performing system initialization, and in fig. 7, the system initialization module 7 includes the following steps when performing system initialization:
step S501 detects a system configuration file: in this step, a system configuration file is detected.
Step S502 determines whether or not it is the first installation: in this step, it is judged whether the first installation is performed, and if the judgment result is yes, step S503 is executed; otherwise, step S506 is executed.
Step S503 installs a database: in this step, a database is installed. After the present step is executed, step S504 is executed.
Step S504 configures a local file: in this step, a local file is configured.
Step S505 initializes a system interface: in this step, a system interface is initialized.
Step S506 displays a system interface: in this step, a system interface is displayed.
Fig. 8 is a flowchart of the system setting module in this embodiment when performing system setting, and in fig. 8, the system setting module 8 includes the following steps when performing system setting:
step S601 displays a setting panel, and determines whether to restore default settings: in this step, a setting panel is displayed, and whether to recover the default setting is judged, if yes, step S606 is executed; otherwise, step S602 is performed. The setting panel is a system setting interface, and a user can perform initialization setting on related parameters of the system on the system setting interface, wherein the initialization setting includes setting of information such as an image saving path, the number of default display retrieval results, the size of a default screenshot and the like.
Step S602 inputs setting information: if the judgment result of the step S601 is no, the present step is executed. In this step, setting information is input. After this step is executed, step S603 is executed.
Step S603 determines whether the input setting information is valid: in this step, it is determined whether the input setting information is valid, and if the determination result is yes, step S604 is executed; otherwise, step S602 is performed.
Step S604 saves the setting information: if the judgment result of the step S603 is yes, the present step is executed. In this step, the setting information is saved.
In a word, when the leather cloth of the raw material is selected by the prior industry products such as bags, clothes and the like, purchasing personnel needs to search for a proper material in the raw material supply market according to the order requirement. This is a time consuming and labor intensive process and often times due to operator negligence the manufacturer does not find suitable raw materials to find alternative leather cloths. The image retrieval system for the leather cloth color card in the embodiment can make the finding work of the leather cloth color card simple and convenient. The buyer only needs to take the leather cloth segment color card, obtain the picture by scanning or photographing, upload the picture obtained by scanning or photographing to the server, then compare the database of the interconnected screen of the leather cloth color card image retrieval system, find the material with the highest similarity with the template color card according to the set conditions, and directly display which supplier has the material sold in the leather cloth color card image retrieval system, so that the buyer has the target to go directly to the corresponding supplier grade, or to each raw material supplier in the supply market where the leather cloth color card image retrieval system is installed, when the buyer asks for the material, the buyer can acquire and retrieve the similar leather cloth color card through the leather cloth color card image retrieval system on the own computer, and the existence of the needed material can be known through the leather cloth color card image retrieval system, or the existence of similar materials solves the problems of difficult material finding, difficult material detection and difficult classification in the field of leather and cloth product processing. It is worth unifying that even if there is no such material, the leather cloth color card image retrieval system records the color card material required by the buyer, and can calculate the heat sales of the material.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (9)
1. A leather cloth color card image retrieval system, comprising:
an image acquisition module: the system comprises a database, a matching precision threshold value and a characteristic extraction plan, wherein the database is used for storing a plurality of visual characteristics of a user;
an image feature extraction module: the system is used for extracting the visual features of the image to be retrieved by combining the selected visual feature extraction method according to the selection combination of the user on the visual features and the selection of the matching precision threshold value, and storing the extracted visual feature multidimensional vector into an image feature library; wherein the selected combination of visual features refers to whether texture features are considered;
an image feature granularity calculation module: the compatibility granularity used for dividing the image by the domain knowledge input by the user and the extracted visual features;
the image retrieval matching module: the image matching method is used for calculating the distance between the image to be retrieved and the characteristics of each image in the image characteristic library according to the selected similarity calculation method, and screening out the images which accord with the matching precision threshold; wherein the similarity calculation method is combined with the concept of compatible particles in particle calculation and the distance between characteristic components to perform measurement; obtaining classification information of the image to be retrieved through extracting domain knowledge of the image to be retrieved and image color texture characteristics; firstly, through division of domain knowledge, obtaining an image which is the same as the material and the processing technology domain knowledge of the leather cloth image to be retrieved, and obtaining a compatible image in a library; then, through the retrieval operation of the color granularity, the image in the library compatible with the color granularity is obtained, and the process comprises the granularity operation of the color dominant color number and the operation of the color containing relation granularity; if the retrieval result is larger than a certain preset threshold value, further dividing by combining the granularity of the image texture features to obtain the compatible granularity of the combined image texture features, and finally obtaining the most similar matching result;
the color card information modification module: the system is used for modifying and storing the color card information in the color card library and verifying and detecting the modified value; the color card information comprises supplier source, price and contact information;
the retrieval result display module: and the system is used for displaying the retrieval result and carrying out secondary screening according to the screening condition of the user.
2. The leather cloth color card image retrieval system of claim 1, further comprising:
a system initialization module: the system is used for initializing all parameters of the leather cloth color card image retrieval system;
a system setting module: modifying various parameters of the leather cloth color card image retrieval system;
an image interception module: the image acquisition device is used for intercepting the size of an image in the image acquisition process;
color card deletion module: the system is used for deleting the color card images individually and in batches;
a character retrieval module: the method is used for searching the images in the color card library through the keyword information.
3. The leather cloth color card image retrieval system of claim 2, wherein the image acquisition module, when performing image acquisition, comprises the steps of:
A) judging whether system import collection or scanning collection is adopted, and if scanning collection is adopted, executing the step B); if the system import collection is adopted, executing the step C);
B) setting scanning parameters and executing the step C); the scanning parameters comprise size, resolution and storage format;
C) displaying the acquired image;
D) judging whether the acquired image is subjected to automatic screenshot or manual screenshot, and if the acquired image is subjected to the manual screenshot, executing the step E); if the screenshot is automatically captured, executing step F);
E) setting scanning parameters again, and executing the step F); the scanning parameters comprise size, resolution and storage format;
F) displaying the screenshot image, and judging whether the screenshot image meets the requirements, if so, executing the step G); otherwise, returning to the step D);
G) judging whether to perform image warehousing or image scanning comparison, and if so, executing the step H); if the image scanning comparison is carried out, executing the step I);
H) inputting image information, and executing the step I); the image information comprises a color card name, a color card number and a supplier;
I) and warehousing the image, and displaying the image after screenshot.
4. The leather cloth color card image retrieval system of claim 1, wherein the image feature extraction module, when performing image feature extraction, comprises the steps of:
A1) carrying out image preprocessing; the image preprocessing comprises gray level histogram equalization, image smoothing and normalization operations;
B1) selecting a feature extraction option and judging whether the texture features are considered, if so, executing the step C1); otherwise, performing step D1);
C1) extracting image texture features, and executing step D1);
D1) extracting image color features;
E1) and normalizing the image features and storing the extracted features into the image feature library.
5. The leather cloth color card image retrieval system according to claim 1, wherein the image retrieval matching module comprises the following steps when performing image retrieval matching:
A2) judging whether the simple retrieval or the advanced retrieval is selected, if so, executing the step B2); if so, perform step C2);
B2) setting a retrieval condition, and executing the step C2); the retrieval conditions comprise image types, retrieval numbers and retrieval precision;
C2) extracting the characteristics of the image to be matched;
D2) performing similarity measurement on the image to be matched and each image in the image feature library, and judging whether secondary screening operation is performed, if so, executing step E2); otherwise, performing step F2);
E2) setting secondary screening conditions, and executing step F2); the secondary screening conditions comprise leather cloth material, supplier and serial number;
F2) and displaying the retrieval result and finishing the retrieval matching operation.
6. The leather cloth color card image retrieval system of claim 1, wherein the retrieval result display module, when displaying the retrieval result, comprises the steps of:
A3) initializing a display panel;
B3) judging whether to select the search result display or the management display, if the search result display is performed, executing step C3); if yes, executing step D3);
C3) acquiring retrieval screening conditions and executing step D3);
D3) displaying the result, and finishing displaying the color card image and the related material information; the related material information comprises supplier source, price and contact information.
7. The leather cloth color card image retrieval system of claim 1, wherein the color card information modification module, when modifying the color card information, comprises the steps of:
A4) opening a color card information modification page, and judging whether to modify the color card which is put into storage or not, if so, executing the step B4); if the warehoused color card is selected to be modified, executing the step C4);
B4) acquiring the existing information of the color cards in the color card library, and executing the step C4);
C4) filling or modifying the color card information and verifying the validity of the filled or modified color card information, if so, executing step D4); otherwise, filling or modifying the color card information again; the color card information comprises supplier source, price and contact information;
D4) and storing the filled or modified color card information into the color card library.
8. The leather cloth color card image retrieval system of claim 2, wherein the system initialization module, when performing system initialization, comprises the steps of:
A5) detecting a system configuration file;
B5) judging whether the installation is the first installation, if so, executing the step C5); otherwise, performing step F5);
C5) installing a database;
D5) configuring a local file;
E5) initializing a system interface;
F5) and displaying a system interface.
9. The leather cloth color card image retrieval system of claim 2, wherein the system setting module, when performing system setting, comprises the steps of:
A6) displaying a setting panel, and judging whether to restore the default setting, if so, executing the step D6); otherwise, performing step B6);
B6) inputting setting information, and executing step C6);
C6) judging whether the input setting information is valid, if so, executing step D6); otherwise, returning to the step B6);
D6) and saving the setting information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510598398.8A CN106547760B (en) | 2015-09-18 | 2015-09-18 | Image retrieval system for leather cloth color card |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510598398.8A CN106547760B (en) | 2015-09-18 | 2015-09-18 | Image retrieval system for leather cloth color card |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106547760A CN106547760A (en) | 2017-03-29 |
| CN106547760B true CN106547760B (en) | 2020-07-31 |
Family
ID=58362496
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510598398.8A Active CN106547760B (en) | 2015-09-18 | 2015-09-18 | Image retrieval system for leather cloth color card |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106547760B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108021603A (en) * | 2017-09-25 | 2018-05-11 | 嘉善锐琦花边有限公司 | A kind of fabrics sample searches apparatus and method based on texture analysis and cloud computing |
| CN108090230A (en) * | 2018-01-10 | 2018-05-29 | 浙江中国轻纺城网络有限公司 | A kind of fabric searching method and device, computer readable storage medium |
| CN109947970B (en) * | 2019-03-16 | 2021-01-15 | 南通联发信息科技有限公司 | Textile fabric pattern retrieval system |
| CN117150089B (en) * | 2023-10-26 | 2023-12-22 | 环球数科集团有限公司 | Character artistic image changing system based on AIGC technology |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4643855B2 (en) * | 2001-05-25 | 2011-03-02 | 関西ペイント株式会社 | Paint combination search system |
| CN104281849B (en) * | 2013-07-03 | 2017-09-19 | 广州盖特软件有限公司 | A kind of cloth color of image feature extracting method |
| CN104281588B (en) * | 2013-07-03 | 2017-10-27 | 广州盖特软件有限公司 | A kind of cloth image search method based on many granularities |
| CN104331447A (en) * | 2014-10-29 | 2015-02-04 | 邱桃荣 | Cloth color card image retrieval method |
-
2015
- 2015-09-18 CN CN201510598398.8A patent/CN106547760B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN106547760A (en) | 2017-03-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11191342B2 (en) | Techniques for identifying skin color in images having uncontrolled lighting conditions | |
| US10049466B2 (en) | Color name generation from images and color palettes | |
| US9396560B2 (en) | Image-based color palette generation | |
| US9552656B2 (en) | Image-based color palette generation | |
| US9401032B1 (en) | Image-based color palette generation | |
| US9311889B1 (en) | Image-based color palette generation | |
| US9135719B1 (en) | Color name generation from images and color palettes | |
| CN106547760B (en) | Image retrieval system for leather cloth color card | |
| CN108228872A (en) | Facial image De-weight method and device, electronic equipment, storage medium, program | |
| CN110472082B (en) | Data processing method, data processing device, storage medium and electronic equipment | |
| US20210174135A1 (en) | Method of matching image and apparatus thereof, device, medium and program product | |
| KR20180131621A (en) | Content-based search and retrieval of brand images | |
| US10769585B2 (en) | Systems and methods for automated harmonized system (HS) code assignment | |
| CN111814828B (en) | Intelligent space planning method, device, equipment and storage medium | |
| CN109740417A (en) | Invoice type recognition methods, device, storage medium and computer equipment | |
| CN112464803A (en) | Image comparison method and device | |
| CN104239873A (en) | Image processing apparatus and processing method | |
| WO2020134411A1 (en) | Merchandise category recognition method, apparatus, and electronic device | |
| CN114730377A (en) | Shoe authentication device and authentication process | |
| US10402777B2 (en) | Method and a system for object recognition | |
| JP7357708B2 (en) | Image matching system | |
| CN108647600B (en) | Face recognition method, face recognition device and computer-readable storage medium | |
| CN113688261B (en) | Image data cleaning method and device, electronic equipment and readable storage medium | |
| CN114359997A (en) | Service guiding method and system | |
| CN111259691B (en) | Biological characteristic analysis method and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |