CN113963348A - Microorganism identification system and method based on Django and image feature matching - Google Patents
Microorganism identification system and method based on Django and image feature matching Download PDFInfo
- Publication number
- CN113963348A CN113963348A CN202111217307.3A CN202111217307A CN113963348A CN 113963348 A CN113963348 A CN 113963348A CN 202111217307 A CN202111217307 A CN 202111217307A CN 113963348 A CN113963348 A CN 113963348A
- Authority
- CN
- China
- Prior art keywords
- microorganism
- django
- image
- feature
- matching
- 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.)
- Withdrawn
Links
- 244000005700 microbiome Species 0.000 title claims abstract description 67
- 238000000034 method Methods 0.000 title claims abstract description 31
- 239000011159 matrix material Substances 0.000 claims description 22
- 239000013598 vector Substances 0.000 claims description 13
- 230000004044 response Effects 0.000 claims description 12
- 230000008569 process Effects 0.000 claims description 10
- 230000008859 change Effects 0.000 claims description 7
- 238000001514 detection method Methods 0.000 claims description 6
- 230000000813 microbial effect Effects 0.000 claims description 6
- 230000000007 visual effect Effects 0.000 abstract description 6
- 238000001000 micrograph Methods 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 9
- 241000196324 Embryophyta Species 0.000 description 3
- 230000008901 benefit Effects 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 241000235349 Ascomycota Species 0.000 description 1
- 210000004102 animal cell Anatomy 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000019771 cognition Effects 0.000 description 1
- 230000001149 cognitive effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- NUHSROFQTUXZQQ-UHFFFAOYSA-N isopentenyl diphosphate Chemical compound CC(=C)CCO[P@](O)(=O)OP(O)(O)=O NUHSROFQTUXZQQ-UHFFFAOYSA-N 0.000 description 1
- 230000004807 localization Effects 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000004806 packaging method and process Methods 0.000 description 1
- 230000002787 reinforcement Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000000844 transformation Methods 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/20—Education
Landscapes
- Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Primary Health Care (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Educational Administration (AREA)
- Marketing (AREA)
- Educational Technology (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a microorganism identification system and method based on Django and image feature matching, which comprises the following steps: module M1: building a microorganism sample library website by using Django and MySQL; module M2: uploading, by a browser, the identified and classified microorganism samples into a sample bank; module M3: and after the user acquires the image by using the electron eyepiece microscope and uploads the image to a website, matching the existing data in the sample library by using the SURF algorithm to finally obtain the microorganism category identity degree list. The invention realizes the automatic identification of common microorganisms in the microscope image information and the display in the visual interface, so that students can know the microorganisms by observing the microorganisms with their own eyes, and compared with the traditional method, the learning method can be accepted by the students more easily and more comprehensively by acquiring the identity information of the microorganisms from the image data.
Description
Technical Field
The invention relates to the technical field of microorganism identification, in particular to a microorganism identification system and method based on Django and image feature matching.
Background
The role of microorganisms as ancient organisms on earth is not insignificant. However, the public cognition on the microorganisms is usually only from class, and the education of the microorganisms is considered to be important by 88.68% of classmates in school investigation. The current teaching mode of microorganism in class has higher requirements on the experience of teachers, but teachers with high teaching experience are few, and new teachers need to learn to save the experience. In the aspect of students, the traditional microscope eyepieces used by the students at present are too small to be convenient for careful observation. Various students have different requirements on knowledge, and children with strong learning ability have strong curiosity and can not meet the classroom knowledge, and can explore unknown microorganisms by using a microscope outside a class, but at the moment, because of lack of guidance of teachers, the students have difficulty in identifying the microorganisms; for children with weak learning ability, the progress of the experiment class may not be followed, the reinforcement needs to be carried out in the class, and the condition that teachers need to be taught is also provided. These difficulties can be solved by a set of microorganism identification system.
Patent document CN100593172C (application number: CN200810093760.6) discloses a microorganism identification system and method based on microscopic images, which utilize computer image processing technology to preprocess the acquired grain storage microorganism microscopic images, automatically extract the mathematical statistics characteristics such as texture and geometric shape of the images according to the target areas of the grain storage microorganism microscopic images, and then use BP neural network to perform classification identification, so as to accurately identify the microorganisms in the grains. However, the application environment of the patent is limited, and the patent cannot be applied to the masses.
Disclosure of Invention
In view of the defects in the prior art, the invention aims to provide a microorganism identification system and method based on Django and image feature matching.
According to the invention, the microorganism identification system based on Django and image feature matching comprises:
module M1: building a microorganism sample library website by using Django and MySQL;
module M2: uploading, by a browser, the identified and classified microorganism samples into a sample bank;
module M3: and after the user acquires the image by using the electron eyepiece microscope and uploads the image to a website, matching the existing data in the sample library by using the SURF algorithm to finally obtain the microorganism category identity degree list.
Preferably, OpenCV is used to build a microorganism sample library, and the following algorithm is deployed:
scale invariant feature transform algorithm SIFT: the system is used for detecting and describing the local characteristics in the image, searching for an extreme point in a spatial scale, and extracting the position, the scale and the rotation invariant of the extreme point;
speedup robust features algorithm SURF: for computer vision tasks, including object recognition and 3D reconstruction.
Preferably, the SURF algorithm is used for feature point detection and matching, and the process is as follows:
constructing a scale space by using Laplace Gaussian difference;
calculating a related Hessian matrix, finding an extreme point of a scale space to determine a characteristic point, constructing a square region of the characteristic point field, wherein the side length is 20 sigma, sigma is the scale of the interest point, and rotating the square region to the main direction of the characteristic point;
dividing the square area into 4 multiplied by 4 sub-areas, and calculating 4-dimensional feature vectors in a 5 multiplied by 5 regular grid space in each sub-area, wherein the features comprise the response of Haar wavelets to the horizontal and vertical directions and the absolute value of the sum of the responses;
and 4-dimensional features of each sub-region are calculated and accumulated, and the divided 16 sub-regions are accumulated to obtain a final 64-dimensional feature vector descriptor.
Preferably, the characteristic point principal direction identification process is as follows: and accumulating the Haar wavelet corresponding values of all points in the sector of 60 degrees in the horizontal and vertical directions in a preset field by taking the characteristic point as a center, wherein the maximum Haar corresponding accumulated value is the main direction corresponding to the characteristic point.
Preferably, the matching degree is determined by calculating the Euclidean distance between the two feature points through the SURF algorithm, and the shorter the Euclidean distance is, the higher the matching degree of the two feature points is;
performing Hessian matrix trace judgment through an SURF algorithm, and if the signs of matrix traces of two characteristic points are the same, representing that the two characteristics have contrast ratio changes in the same direction; if the contrast ratio is different, the contrast ratio change directions of the two characteristic points are opposite and are directly excluded.
The microorganism identification method based on Django and image feature matching provided by the invention comprises the following steps:
step 1: building a microorganism sample library website by using Django and MySQL;
step 2: uploading, by a browser, the identified and classified microorganism samples into a sample bank;
and step 3: and after the user acquires the image by using the electron eyepiece microscope and uploads the image to a website, matching the existing data in the sample library by using the SURF algorithm to finally obtain the microorganism category identity degree list.
Preferably, OpenCV is used to build a microorganism sample library, and the following algorithm is deployed:
scale invariant feature transform algorithm SIFT: the system is used for detecting and describing the local characteristics in the image, searching for an extreme point in a spatial scale, and extracting the position, the scale and the rotation invariant of the extreme point;
speedup robust features algorithm SURF: for computer vision tasks, including object recognition and 3D reconstruction.
Preferably, the SURF algorithm is used for feature point detection and matching, and the process is as follows:
constructing a scale space by using Laplace Gaussian difference;
calculating a related Hessian matrix, finding an extreme point of a scale space to determine a characteristic point, constructing a square region of the characteristic point field, wherein the side length is 20 sigma, sigma is the scale of the interest point, and rotating the square region to the main direction of the characteristic point;
dividing the square area into 4 multiplied by 4 sub-areas, and calculating 4-dimensional feature vectors in a 5 multiplied by 5 regular grid space in each sub-area, wherein the features comprise the response of Haar wavelets to the horizontal and vertical directions and the absolute value of the sum of the responses;
and 4-dimensional features of each sub-region are calculated and accumulated, and the divided 16 sub-regions are accumulated to obtain a final 64-dimensional feature vector descriptor.
Preferably, the characteristic point principal direction identification process is as follows: and accumulating the Haar wavelet corresponding values of all points in the sector of 60 degrees in the horizontal and vertical directions in a preset field by taking the characteristic point as a center, wherein the maximum Haar corresponding accumulated value is the main direction corresponding to the characteristic point.
Preferably, the matching degree is determined by calculating the Euclidean distance between the two feature points through the SURF algorithm, and the shorter the Euclidean distance is, the higher the matching degree of the two feature points is;
performing Hessian matrix trace judgment through an SURF algorithm, and if the signs of matrix traces of two characteristic points are the same, representing that the two characteristics have contrast ratio changes in the same direction; if the contrast ratio is different, the contrast ratio change directions of the two characteristic points are opposite and are directly excluded.
Compared with the prior art, the invention has the following beneficial effects:
(1) the system aims to help students to independently learn and distinguish microorganisms, and in the process of exploring a cognitive microorganism world, the interest in microorganism or biological research is cultivated, the system realizes automatic identification of common microorganisms in microscope image information and display in a visual interface, and the students can recognize the microorganisms by self-eye observation more, so that the learning method is easier for the students to accept and more comprehensive than the traditional method such as acquiring the identity information of the microorganisms from picture data;
(2) the system can be applied to student microbial experiment courses, on one hand, students can be helped to learn and consolidate knowledge granted by courses, on the other hand, teachers can be assisted to give courses, the teaching difficulty is reduced, the system can also be used as a tool for science popularization education, students can know the microbial knowledge nearby by themselves, and the vast microbial world is explored;
(3) the system can be used by installing the software system only on the basis of the original electronic eyepiece in a laboratory without additional hardware expenditure.
Drawings
Other features, objects and advantages of the invention will become more apparent upon reading of the detailed description of non-limiting embodiments with reference to the following drawings:
FIG. 1 is a schematic scale space diagram of the SURF algorithm;
FIG. 2 is a scale space diagram of SIFT algorithm;
FIG. 3 is a schematic diagram of feature point locations;
FIG. 4 is a Microbe table;
FIG. 5 is a Category table;
FIG. 6 is a system interface diagram;
FIG. 7 is a hardware diagram of the whole system;
FIG. 8 is a process flow diagram;
fig. 9 is a frame diagram of the working principle.
Detailed Description
The present invention will be described in detail with reference to specific examples. The following examples will assist those skilled in the art in further understanding the invention, but are not intended to limit the invention in any way. It should be noted that it would be obvious to those skilled in the art that various changes and modifications can be made without departing from the spirit of the invention. All falling within the scope of the present invention.
Example (b):
according to the invention, the microorganism identification system based on Django and image feature matching comprises: a microorganism sample library website is built by using Django and MySQL, a user can upload identified and classified microorganism samples to a system database through a browser, after the user obtains an image by using an electronic eyepiece microscope, the image can be uploaded through the website, a system background matches existing data in the library by using an SURF algorithm, and finally a microorganism category identity degree list is obtained.
The method is characterized in that OpenCV is selected when a microbial database is built, and an SURF algorithm which is high in speed and accurate in identification is selected on a local feature algorithm, is an improvement on SIFT, and is mainly characterized by being fast. In addition, on a visual interface, a Django-based website form is selected, and deployment and use are facilitated.
The OpenCV is: since OpenCV is a real-time application dedicated to the real world, the execution speed of the OpenCV is greatly improved by writing optimized C code, and since the matching microorganism species are a huge database, a light-weight and efficient visual library of OpenCV is selected in the aspect of the visual library. Also, if intel's processor is used, OpenCV automatically calls IPPICV. IPPICV can be linked to OpenCV at the compile stage, thus replacing the corresponding low-level optimized C language code, and the speed increase obtained using IPP is considerable.
The Django: in order to enable the application to be better integrated into the public, the software specially applies a Django Web page development framework which is written by using Python language, is an open-source Web framework and uses an MVC design mode like other Web frameworks. And Django is very convenient relative to other development platforms, and has the greatest advantage of packaging, processing and mining data. In addition, Django can be seamlessly matched with OpenCV, and the complexity of the system is reduced.
The local feature algorithm:
(1) scale-invariant feature transform (SIFT) is an algorithm of computer vision for detecting and describing local features in an image, and is used for searching extreme points in a spatial Scale and extracting position, Scale and rotation invariants of the extreme points;
(2) SURF (Speeded Up Robust Features) is a Robust image recognition and description algorithm that can be used for computer vision tasks such as object recognition and 3D reconstruction. The inspiration of his part comes from the SIFT algorithm. The SURF standard is several times faster in version than SIFT and its authors claim to be more robust than SIFT in terms of different image transformations.
The SURF algorithm is obtained by reducing descriptor dimensionality through simplifying a Gaussian template on the basis of SIFT, and the running speed is 2-3 times faster than that of SIFT. SIFT is more stable at larger rotations, but SURF is less different from SIFT in stability but SURF is more efficient with only translation of the microscopic image. Compared with Harris and Susan isocenter detection operators, the SURF has the advantages of illumination, rotation invariance and the like, and the extracted feature points are more stable, so that the SURF algorithm is used for feature point detection and matching in the text. The method comprises the following specific steps:
step 1: constructing a scale space by using Laplace Gaussian difference;
step 2: and detecting key points. Calculating a relevant Hessian matrix and finding a scale space extreme point to determine a key point, wherein the Hessian matrix is defined as follows:
and step 3: the keypoint direction is assigned. And taking the characteristic point as a center, accumulating the Haar wavelet corresponding values of all points in the sector of 60 degrees in the horizontal and vertical directions in a certain field, wherein the maximum Haar corresponding accumulated value is the main direction corresponding to the characteristic point.
And 4, step 4: and generating a characteristic point descriptor.
Step 4.1: constructing a square region of the characteristic point field, wherein the side length is 20 sigma, sigma is the scale of an interest point, and rotating the square region to the main direction of the characteristic point;
step 4.2: dividing the square area established in the last step into 4 multiplied by 4 sub-areas, and calculating 4-dimensional feature vectors in a 5 multiplied by 5 regular grid space in each sub-area, wherein the features comprise the response of the Haar wavelet to the horizontal direction and the vertical direction and the absolute value of the sum of the responses;
step 4.3: and 4-dimensional features of each sub-region are calculated and accumulated, and the divided 16 sub-regions are accumulated to obtain a final 64-dimensional feature vector descriptor.
wxPython is a source opening software, has good compatibility with cross-platform, and can support operation under Windows, most Unix systems and Macintosh OSX. And the language of Python is simple and concise, so when the visual interface is written, the wx framework under the Python language is selected. The whole set of visualization program is written out using wxPython.
Algorithm selection, as in fig. 1 and 2:
(1) in the aspect of generating scale space, the SIFT algorithm utilizes the convolution generation of a difference gaussian pyramid and spatial images of different levels, and the SURF algorithm adopts the convolution of box filters of different scales with an original image.
(2) During feature point inspection, the SIFT operator firstly inhibits the non-maximum value of an image, then removes points with low contrast, and then removes points at the edge through a Hessian matrix; the SURF algorithm firstly detects candidate feature points through a Hessian matrix and then suppresses points with non-maximum values.
(3) In the aspect of determining the direction of the feature vector, the SIFT algorithm is to count a histogram of the amplitude of the gradient in a square region, find the direction corresponding to the maximum gradient amplitude, and the feature points determined by the SIFT operator can have one or more directions, wherein the one or more directions include a main direction and a plurality of auxiliary directions; the SURF algorithm is that in a circular neighborhood, Haar wavelet responses in horizontal and vertical directions in each sector range are detected, the sector direction with the maximum module value is found, and the algorithm has only one direction.
(4) When the SIFT algorithm generates the descriptors, 16 × 16 sampling points are divided into 4 × 4 regions, so that the amplitude of each division seed point is calculated and the direction of the division seed point is determined, and the total number is 4 × 8 to 128 dimensions.
The SURF algorithm, when generating the feature descriptors, divides a 20 × 20s square into 4 × 4 small squares, 25 samples per sub-region, calculates the wavelet haar response v [ Σdx, Σ dy, Σ | dx |, ∑ dy | ], and has a total of 4 × 4 ═ 64 dimensions.
To sum up, the SURF algorithm is finally selected. Because the SURF algorithm simplifies some tedious work in each step, only one main direction of the feature points is calculated, and the dimension of the generated feature descriptor is reduced compared with the former.
The algorithm is realized as follows:
(1) construction of Hessian matrix
SURF uses a Hessian matrix determinant approximation image. The Hessian matrix is constructed to generate stable edge points (catastrophe points) of the image, so that the feature extraction is faster.
(2) Feature point localization
As shown in fig. 3, each pixel point processed by the Hessian matrix is compared with 26 points in the neighborhood of the two-dimensional image space and the scale space, the key point is preliminarily located, and the final stable feature point is screened out by filtering the key point with weaker energy and the key point with wrong location. And then adopting harr wavelet characteristics in the circular neighborhood of the statistical characteristic points. As the principal direction of the feature point.
(3) Characteristic point descriptor
In the SURF algorithm, one rectangular region block 44 is also taken around the feature point, but the direction of the taken rectangular region is along the main direction of the feature point. Each subregion counts haar wavelet features of 25 pixels in both the horizontal and vertical directions, where both the horizontal and vertical directions are relative to the principal direction. The haar wavelet features are 4 directions of the sum of the horizontal direction value, the vertical direction value, the horizontal direction absolute value and the vertical direction absolute value. These 4 values are used as the feature vector of each sub-block region, so a total 4 × 4 — 64-dimensional vector is used as the descriptor of SURF features, which is reduced by half compared with the descriptor of SIFT features.
(4) Descriptor matching
SURF determines the degree of matching by calculating the euclidean distance between two feature points, the shorter the euclidean distance, the better the degree of matching representing the two feature points. SURF also uses the judgment of Hessian matrix trace, if the signs of the matrix traces of two characteristic points are the same, the two characteristic points have contrast change in the same direction, if the signs are different, the contrast change directions of the two characteristic points are opposite, and even if the Euclidean distance is 0, the contrast change directions are directly excluded.
Specimen information base
The Microbe table is shown in FIG. 4, the Category table is shown in FIG. 5, and FIGS. 4 and 5 are descriptions of information on the microbial specimens recorded in the library. When the program identifies the corresponding specimen, it is called from the library. At present, 6 kinds of microorganism specimen information are recorded in a library and belong to 5 categories. Is a dicotyledonous plant, a herbaceous plant, a higher plant, an animal cell and an ascomycete respectively. Future extensions enter more sample data.
Fig. 6 is a system interface diagram, in which the upper left part displays the classification of microorganisms, so as to facilitate the user to browse and screen. At the bottom left are two buttons that allow the user to upload and identify microbes. All the stored microorganism pictures and descriptions in the library are displayed on the right. Fig. 7 is a hardware diagram of the whole system. Hardware aspect: one microscope with an electronic eyepiece and one computer. Software aspect: an image feature matching based microscopic microorganism identification system.
Fig. 8 and 9 show the working principle and flow of the system of the present invention: the electronic eyepiece which can be purchased in the market is connected with a computer USB, after the connection is completed, the software of people is opened, after a real-time image is seen, the glass slide is placed under a microscope, people judge whether focusing is completed or not through the software real-time image, and after the focusing is completed, an image is obtained. The system will return the result by uploading the image just obtained through the 'identify my microorganism' button on the website.
Those skilled in the art will appreciate that, in addition to implementing the systems, apparatus, and various modules thereof provided by the present invention in purely computer readable program code, the same procedures can be implemented entirely by logically programming method steps such that the systems, apparatus, and various modules thereof are provided in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Therefore, the system, the device and the modules thereof provided by the present invention can be considered as a hardware component, and the modules included in the system, the device and the modules thereof for implementing various programs can also be considered as structures in the hardware component; modules for performing various functions may also be considered to be both software programs for performing the methods and structures within hardware components.
The foregoing description of specific embodiments of the present invention has been presented. It is to be understood that the present invention is not limited to the specific embodiments described above, and that various changes or modifications may be made by one skilled in the art within the scope of the appended claims without departing from the spirit of the invention. The embodiments and features of the embodiments of the present application may be combined with each other arbitrarily without conflict.
Claims (10)
1. A microorganism identification system based on Django and image feature matching, comprising:
module M1: building a microorganism sample library website by using Django and MySQL;
module M2: uploading, by a browser, the identified and classified microorganism samples into a sample bank;
module M3: and after the user acquires the image by using the electron eyepiece microscope and uploads the image to a website, matching the existing data in the sample library by using the SURF algorithm to finally obtain the microorganism category identity degree list.
2. The microorganism identification system based on Django and image feature matching according to claim 1, wherein OpenCV is used to construct a microorganism sample library and deploy the following algorithm:
scale invariant feature transform algorithm SIFT: the system is used for detecting and describing the local characteristics in the image, searching for an extreme point in a spatial scale, and extracting the position, the scale and the rotation invariant of the extreme point;
speedup robust features algorithm SURF: for computer vision tasks, including object recognition and 3D reconstruction.
3. The microorganism identification system based on Django and image feature matching according to claim 2, wherein the SURF algorithm is used for feature point detection and matching, and the process is as follows:
constructing a scale space by using Laplace Gaussian difference;
calculating a related Hessian matrix, finding an extreme point of a scale space to determine a characteristic point, constructing a square region of the characteristic point field, wherein the side length is 20 sigma, sigma is the scale of the interest point, and rotating the square region to the main direction of the characteristic point;
dividing the square area into 4 multiplied by 4 sub-areas, and calculating 4-dimensional feature vectors in a 5 multiplied by 5 regular grid space in each sub-area, wherein the features comprise the response of Haar wavelets to the horizontal and vertical directions and the absolute value of the sum of the responses;
and 4-dimensional features of each sub-region are calculated and accumulated, and the divided 16 sub-regions are accumulated to obtain a final 64-dimensional feature vector descriptor.
4. The microorganism identification system based on Django and image feature matching according to claim 3, wherein the feature point principal direction identification process is as follows: and accumulating the Haar wavelet corresponding values of all points in the sector of 60 degrees in the horizontal and vertical directions in a preset field by taking the characteristic point as a center, wherein the maximum Haar corresponding accumulated value is the main direction corresponding to the characteristic point.
5. The microbial identification system based on Django and image feature matching of claim 3, wherein the matching degree is determined by calculating Euclidean distance between two feature points through SURF algorithm, the shorter the Euclidean distance is, the higher the matching degree of the two feature points is represented;
performing Hessian matrix trace judgment through an SURF algorithm, and if the signs of matrix traces of two characteristic points are the same, representing that the two characteristics have contrast ratio changes in the same direction; if the contrast ratio is different, the contrast ratio change directions of the two characteristic points are opposite and are directly excluded.
6. A microorganism identification method based on Django and image feature matching is characterized by comprising the following steps:
step 1: building a microorganism sample library website by using Django and MySQL;
step 2: uploading, by a browser, the identified and classified microorganism samples into a sample bank;
and step 3: and after the user acquires the image by using the electron eyepiece microscope and uploads the image to a website, matching the existing data in the sample library by using the SURF algorithm to finally obtain the microorganism category identity degree list.
7. The microorganism identification method based on Django and image feature matching as claimed in claim 6, wherein OpenCV is used to build a microorganism sample library and deploy the following algorithm:
scale invariant feature transform algorithm SIFT: the system is used for detecting and describing the local characteristics in the image, searching for an extreme point in a spatial scale, and extracting the position, the scale and the rotation invariant of the extreme point;
speedup robust features algorithm SURF: for computer vision tasks, including object recognition and 3D reconstruction.
8. The microorganism identification method based on Django and image feature matching according to claim 7, wherein the SURF algorithm is used for feature point detection and matching, and the process is as follows:
constructing a scale space by using Laplace Gaussian difference;
calculating a related Hessian matrix, finding an extreme point of a scale space to determine a characteristic point, constructing a square region of the characteristic point field, wherein the side length is 20 sigma, sigma is the scale of the interest point, and rotating the square region to the main direction of the characteristic point;
dividing the square area into 4 multiplied by 4 sub-areas, and calculating 4-dimensional feature vectors in a 5 multiplied by 5 regular grid space in each sub-area, wherein the features comprise the response of Haar wavelets to the horizontal and vertical directions and the absolute value of the sum of the responses;
and 4-dimensional features of each sub-region are calculated and accumulated, and the divided 16 sub-regions are accumulated to obtain a final 64-dimensional feature vector descriptor.
9. The microorganism identification method based on Django and image feature matching according to claim 8, wherein the feature point principal direction identification process is as follows: and accumulating the Haar wavelet corresponding values of all points in the sector of 60 degrees in the horizontal and vertical directions in a preset field by taking the characteristic point as a center, wherein the maximum Haar corresponding accumulated value is the main direction corresponding to the characteristic point.
10. The microorganism identification method based on Django and image feature matching according to claim 8, wherein the matching degree is determined by calculating Euclidean distance between two feature points through SURF algorithm, the shorter the Euclidean distance is, the higher the matching degree of the two feature points is represented;
performing Hessian matrix trace judgment through an SURF algorithm, and if the signs of matrix traces of two characteristic points are the same, representing that the two characteristics have contrast ratio changes in the same direction; if the contrast ratio is different, the contrast ratio change directions of the two characteristic points are opposite and are directly excluded.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111217307.3A CN113963348A (en) | 2021-10-19 | 2021-10-19 | Microorganism identification system and method based on Django and image feature matching |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111217307.3A CN113963348A (en) | 2021-10-19 | 2021-10-19 | Microorganism identification system and method based on Django and image feature matching |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113963348A true CN113963348A (en) | 2022-01-21 |
Family
ID=79465483
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111217307.3A Withdrawn CN113963348A (en) | 2021-10-19 | 2021-10-19 | Microorganism identification system and method based on Django and image feature matching |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113963348A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114724142A (en) * | 2022-04-02 | 2022-07-08 | 四川大学 | Sewage treatment indicative microorganism image identification method based on convolutional neural network |
-
2021
- 2021-10-19 CN CN202111217307.3A patent/CN113963348A/en not_active Withdrawn
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114724142A (en) * | 2022-04-02 | 2022-07-08 | 四川大学 | Sewage treatment indicative microorganism image identification method based on convolutional neural network |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111448582B (en) | System and method for single channel whole cell segmentation | |
Sadeghi-Tehran et al. | Automated method to determine two critical growth stages of wheat: heading and flowering | |
US20190355113A1 (en) | Multi-sample Whole Slide Image Processing in Digital Pathology via Multi-resolution Registration and Machine Learning | |
DE102017100396A1 (en) | Classifiers for high and low fidelity used on street scene images | |
CN113034495B (en) | Spine image segmentation method, medium and electronic device | |
CN112686850B (en) | Method and system for few-sample segmentation of CT image based on spatial position and prototype network | |
Selda et al. | Plant identification by image processing of leaf veins | |
CN114092938B (en) | Image recognition processing method and device, electronic equipment and storage medium | |
CN113128335A (en) | Method, system and application for detecting, classifying and discovering micro-body paleontological fossil image | |
CN115205727A (en) | Experiment intelligent scoring method and system based on unsupervised learning | |
Dias et al. | Soundscape segregation based on visual analysis and discriminating features | |
CN106650811A (en) | Hyperspectral mixed pixel classification method based on neighbor cooperation enhancement | |
Bonačić et al. | Optical character recognition of seven-segment display digits using neural networks | |
CN113963348A (en) | Microorganism identification system and method based on Django and image feature matching | |
Asmara et al. | Identification of Toga Plants Based on Leaf Image Using the Invariant Moment and Edge Detection Features | |
CN117095270A (en) | Visual recognition method of handwriting identifier based on target detection neural network | |
Chavan et al. | Plant species image recognition using artificial intelligence on jetson nano computational platform | |
Liu | Research on the teaching mode of information visualization course for visual communication design major based on artificial intelligence technology | |
CN105005788B (en) | The target apperception method of simulated human Low Level Vision | |
Okoń et al. | Possibilities for the use of edge detection algorithms in the analysis of images of oilseed rape leaves | |
Largo et al. | Bubble sheet multiple choice mobile checker with test grader using optical mark recognition (OMR) algorithm | |
CN112802049B (en) | Method and system for constructing household article detection data set | |
Mortensen et al. | Pattern recognition for ecological science and environmental monitoring: An initial report | |
Marée et al. | Towards generic image classification: an extensive empirical study | |
Zheng et al. | Localization and recognition of single particle image in microscopy micrographs based on region based convolutional neural networks |
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 | ||
WW01 | Invention patent application withdrawn after publication | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20220121 |