CN111651268B - Quick processing system for microscopic image - Google Patents
Quick processing system for microscopic image Download PDFInfo
- Publication number
- CN111651268B CN111651268B CN202010407145.9A CN202010407145A CN111651268B CN 111651268 B CN111651268 B CN 111651268B CN 202010407145 A CN202010407145 A CN 202010407145A CN 111651268 B CN111651268 B CN 111651268B
- Authority
- CN
- China
- Prior art keywords
- microscopic image
- terminal
- microscopic
- module
- processing
- 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
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
- G06F9/505—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering the load
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/94—Hardware or software architectures specially adapted for image or video understanding
- G06V10/95—Hardware or software architectures specially adapted for image or video understanding structured as a network, e.g. client-server architectures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/69—Microscopic objects, e.g. biological cells or cellular parts
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/69—Microscopic objects, e.g. biological cells or cellular parts
- G06V20/695—Preprocessing, e.g. image segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/03—Recognition of patterns in medical or anatomical images
Abstract
The invention provides a microscopic image rapid processing system, which comprises a terminal and a server, wherein the terminal comprises a microscopic image scanning module, a terminal self-evaluation module, a microscopic image sorting module and a microscopic image processing module, and the terminal evaluates self computing resources, or uploads a scanned microscopic image to the server after cutting, or carries out the work of microscopic image splicing and microscopic image identification on the terminal, so that the computing resources of the terminal are utilized to the maximum extent, and the microscopic image processing speed is greatly improved.
Description
Technical Field
The invention relates to the field of image processing, in particular to a microscopic image rapid processing system.
Background
When a microscopic image acquisition device such as a mobile phone is used for acquiring microscopic images, the acquired microscopic images are too large and too many, for example, each image acquired by the mobile phone is compressed to be 10MB, and 1200 images of a slide can reach 12 GB. Therefore, more service capacity needs to be provided under the same server power resource condition. With the development of terminal equipment hardware in recent years, computing resources are remarkably improved, and for terminal equipment with higher performance, if the microscopic image is subjected to grading processing according to terminal computing resource evaluation, the microscopic image processing rate can be greatly improved.
Chinese patent CN105957008B, "panoramic image real-time stitching method and system based on mobile terminal", performs feature point detection on a current frame of an acquired video stream, performs feature point matching on an obtained feature point attribute sequence of the current frame and a feature point attribute sequence of a previous frame of the video stream, calculates a transformation matrix between the current frame and the previous frame according to a matching result, and finally performs real-time stitching on the current frame and the previous frame according to the transformation matrix, thereby implementing image stitching at the mobile terminal. Chinese patent CN105931187B "image processing method and apparatus" performs thumbnail processing on an image and replaces the original image, so as to save storage space. The technology realizes the splicing or abbreviating processing of the mobile terminal pictures, but lacks the calculation evaluation of the terminal equipment.
Disclosure of Invention
The invention aims to solve the technical problem of providing a microscopic image rapid processing system which can carry out different processing on a microscopic image according to the size of a terminal computational resource and solve the problem of low processing efficiency of the microscopic image.
In order to solve the technical problems, the technical scheme adopted by the invention is as follows: the microscopic image rapid processing system comprises a terminal and a server, wherein the terminal comprises a microscopic image scanning module, a terminal self-evaluation module, a microscopic image sorting module and a microscopic image processing module;
the microscopic image scanning module is used for acquiring microscopic images;
the terminal self-evaluation module is used for evaluating computing resources of the terminal;
the microscopic image sorting module is used for sorting the microscopic images;
the microscopic image processing module is used for carrying out different processing on the microscopic image according to the terminal computing resources evaluated by the terminal self-evaluation module;
the method comprises the following specific steps:
s1, when a terminal operates a microscopic image scanning module for the first time, taking the maximum value of computing power resources required by the microscopic image scanning module as a reserved resource A, and setting the microscopic image scanning module as the highest priority;
s2, scanning by a microscopic image scanning module to obtain a plurality of microscopic images, and sequencing the plurality of microscopic images by a microscopic image sequencing module;
s3, the terminal self-evaluation module terminal performs self-evaluation on the self-calculation capacity of the terminal to obtain a terminal evaluation capacity B;
s4, calculating a difference value between the evaluation value A and the reserved resource B to obtain an available resource C;
and S5, judging by the microscopic image processing module according to the calculation capacity of the available resource C or the client grade, and carrying out terminal operation.
In a preferred scheme, the microscopic image processing module comprises microscopic image cutting, microscopic image splicing and microscopic image identification;
the step S5 is realized by the following steps:
sa1, when available resource C is smaller than computational resource required by microscopic image splicing or the level of a client is high, the microscopic image processing module only performs microscopic image cutting processing, and the cut microscopic image is sent to a terminal workstation or a preprocessing server;
sa2, when the available resource C is larger than the computational resource required by the microscopic image cutting and splicing processing and is smaller than the computational resource required by the microscopic image cutting, the microscopic image splicing and the microscopic image identification processing, the microscopic image processing module performs the microscopic image cutting and the microscopic image splicing processing, and the spliced microscopic image is sent to the server side;
and Sa3, when the available resource C is larger than the computing resource required by the microscopic image cutting, the microscopic image splicing and the microscopic image identification processing, the microscopic image processing module performs the microscopic image cutting, the microscopic image splicing and the microscopic image identification processing, displays the microscopic image identification result and sends the microscopic image identification result to the server side.
In a preferred scheme, the microscopic image sorting module in the step S2 sorts the microscopic images, that is, reads the control command, adds the control commands of the X and Y axes to the numbers of the microscopic images, and obtains the microscopic image matrix sorting through the numbers.
In a preferred scheme, in step Sa1, the customer class is classified to collect a busy traffic level of the customer or an area where the customer is located, and the customer class is set to be high when the traffic is busy or the distance from the server is far, otherwise, the customer class is low.
Preferably, in the step Sa1, the microscopic image is clipped to 128 × 128 to 1024 × 1024 pixels.
In the preferred scheme, in the step Sa2, the microscopic image stitching processes whether the diagonal pixel blocks are adjacent, then fits a spline curve of the picture profile and optimizes control points of the spline curve, and realizes the maximum smoothness and the maximum jagging of the stitched microscopic image on the basis of ensuring accurate identification.
In the preferred scheme, a terminal workstation is set at a high customer level and is connected with a terminal through a high-speed local area network to process microscopic image splicing and microscopic image identification;
the terminal workstation is a multi-core server.
In the preferred scheme, a preprocessing server is set at a high client level, and the incomplete microscopic image splicing and microscopic image identification processing of the terminal in a high bandwidth range are preferentially processed and then sent to a server side.
In a preferred embodiment, the microscope image is stored in jpg, tif or png format.
In a preferred scheme, before the microscopic image scanning module is operated in step S1, it is determined whether a terminal workstation or a preprocessing server exists, if so, only microscopic image cropping is performed, the cropped microscopic image is uploaded to the terminal workstation or the preprocessing server to perform microscopic image stitching and microscopic image recognition, and if not, the microscopic image scanning module is operated to execute the subsequent steps.
The invention provides a microscopic image rapid processing system, which has the following beneficial effects by adopting the scheme:
1. the terminal utilizes the terminal resources to the maximum extent, the terminal cuts the microscopic image and uploads the microscopic image to the server, or the self computing power is evaluated, and the microscopic image splicing or microscopic image recognition work is carried out according to the size of the self computing power resources, so that the computing power resources of the terminal are utilized to the maximum extent.
2. The system greatly improves the processing speed of the microscopic image, transfers partial work of the microscopic image processing from the server side to the terminal, grades the client, only cuts the microscopic image for the busy client, and sets a terminal workstation or a preprocessing server, thereby greatly relieving the pressure of the server and obviously improving the processing speed of the microscopic image.
3. The application range is expanded, and the disease screening can be efficiently and accurately carried out through cell identification in remote areas, thereby being beneficial to integrally improving the disease prevention and control level.
Drawings
The invention is further illustrated with reference to the following figures and examples:
FIG. 1 is a schematic view of the overall process of the present invention;
FIG. 2 is a schematic diagram of the connection of modules in the present invention;
FIG. 3 is a schematic view illustrating the process of matching view sub-blocks according to the present invention;
FIG. 4 is a schematic view of a process for fitting the position of the field of view according to the present invention;
FIG. 5 is a schematic flow chart of block extraction in the present invention;
FIG. 6 is a schematic diagram of the present invention.
Detailed Description
Example 1:
as shown in fig. 1 to 2, the microscopic image fast processing system comprises a terminal and a server, wherein the terminal comprises a microscopic image scanning module, a terminal self-evaluation module, a microscopic image sorting module and a microscopic image processing module;
the specific implementation steps are as follows:
s1, when a terminal operates a microscopic image scanning module for the first time, taking the maximum value of computing resources required by the microscopic image scanning module as reserved resources A, wherein the computing resources comprise CPU (Central processing Unit) proportion resources, memory proportion resources, external memory proportion resources and network bandwidth proportion resources, for example, when a mobile phone terminal operates the system, the CPU usage of each core can be extracted from idle data by reading/proc/stat data, and the CPU usage of each core can be calculated by "/proc/" + pid + "/stat", where a series of calculations are carried out and a CPU time slice of a process is taken. Thereby obtaining CPU proportion resource, memory proportion resource, external memory proportion resource and network bandwidth proportion resource needed by operating the microscopic image scanning module. And the microscopic image scanning module is set to be the highest priority, and the optimal operation of the microscopic image scanning module is ensured in the operation process of the system.
S2, scanning by a microscopic image scanning module to obtain a plurality of microscopic images, wherein the storage format of the microscopic images is jpg, tif or png preferentially, and sequencing the plurality of microscopic images by a microscopic image sequencing module; in the step S2, the microscopic image sorting module sorts the microscopic images, that is, reads the control command, adds the control command of the X and Y axes to the number of the microscopic image, and obtains the microscopic image matrix sorting by the number, for example, the first column of the first row is 0a01, the last column is 0a40, the starting column of the next row is 0B40, and the ending column is 0B01, so that the image matrix sorting can be conveniently obtained by the number, thereby avoiding the need of sorting identification, and reducing the computational resource consumption.
And S3, the terminal self-evaluation module terminal performs self-evaluation on the terminal self-calculation capacity to obtain a terminal evaluation capacity B, for example, the mobile phone terminal can obtain the terminal evaluation capacity B by acquiring the model of the mobile phone and inquiring factory data, and can also adopt manual entry during installation.
S4, calculating a difference value between the evaluation value A and the reserved resource B to obtain an available resource C;
s5, the microscopic image processing module judges according to the computing power of the available resource C, as shown in FIG. 6, taking a processor of high-pass and apple as an example, when the adopted processors are cellcept 730, cellcept 665, cellcept 660, A9, A8 and below, and the flash memory adopts EMMC5.1 and below, and runs the memory 4G and below, namely the available resource C is obviously smaller than the computing power resource required by the microscopic image splicing, the execution level is 3, the microscopic image processing module only carries out the microscopic image cutting processing, firstly, the inSampleSize is set to be n times of the image size expected to be obtained, for example, the image of the original image size 1/4 is expected to be obtained, the value of the inSampleSize is set to be 2, and the images of the original image size 1/2 can be obtained firstly. Then the size of the image is accurately required by setting the attribute of inDensity and inTargetDensity, the image is processed by using a filter, and the microscopic image is cut into 128 × 128 to 1024 × 1024 pixels. The cut microscopic image is sent to a terminal workstation, the terminal workstation is connected with a terminal through a high-speed local area network, the microscopic image splicing and microscopic image recognition are processed, the transmission rate of the microscopic image can be greatly improved, meanwhile, the pressure of a server end is relieved, and the terminal workstation is a multi-core server. When the adopted processors are cellcept 820, cellcept 810, cellcept 765, A10 and the like, the flash memory adopts UFS 2.0, UFS 2.1 and the like, and the memory 6G is operated, namely the available resource C is larger than the computing power resource required by the microscopic image cutting and splicing processing and is smaller than the computing power resource required by the microscopic image cutting, microscopic image splicing and microscopic image identification processing, the machine body capacity execution level is 2 levels, and the microscopic image cutting and microscopic image splicing operation is executed at the terminal. When the adopted processors are cellcept 865, cellcept 845, cellcept 835, A13, A11 and above, the flash memory adopts UFS 3.0, UFS 3.1 and NVMe and above, and the processors with 5G modules are preferably adopted, namely the available resource C is larger than the computing power resource required by the microscopic image cutting, the microscopic image splicing and the microscopic image identification processing, the execution grade is 1 grade, and the terminal can execute the operations of the microscopic image cutting, the microscopic image splicing and the microscopic image identification.
Example 2:
as shown in fig. 1 to 2, the microscopic image fast processing system comprises a terminal and a server, wherein the terminal comprises a microscopic image scanning module, a terminal self-evaluation module, a microscopic image sorting module and a microscopic image processing module;
the specific implementation steps are as follows:
s1, before a microscopic image scanning module is operated, judging whether a terminal workstation or a preprocessing server exists, if so, only performing microscopic image cutting processing, uploading a cut microscopic image to the terminal workstation or the preprocessing server to perform microscopic image splicing and microscopic image identification processing, if not, operating the microscopic image scanning module, taking the maximum value of computing power resources required by the microscopic image scanning module as reserved resources A, wherein the computing power resources comprise CPU (central processing unit) proportion resources, internal memory proportion resources, external memory proportion resources and network bandwidth proportion resources, for example, when a mobile phone terminal operates the system, the CPU usage of each core and idle data can be extracted by reading/proc/stat data, and CPU time slices of a process can be calculated by "/proc/" + pid + "/stat". Therefore, CPU proportion resources, internal memory proportion resources, external memory proportion resources and network bandwidth proportion resources required by the operation of the microscopic image scanning module are obtained. And the microscopic image scanning module is set to be the highest priority, and the optimal operation of the microscopic image scanning module is ensured in the operation process of the system.
S2, scanning by a microscopic image scanning module to obtain a plurality of microscopic images, wherein the storage format of the microscopic images is jpg, tif or png preferentially, and sequencing the plurality of microscopic images by a microscopic image sequencing module; in the step S2, the microscopic image sorting module sorts the microscopic images, that is, reads the control command, adds the control command of the X and Y axes to the number of the microscopic image, and obtains the microscopic image matrix sorting by the number, for example, the first column of the first row is 0a01, the last column is 0a40, the starting column of the next row is 0B40, and the ending column is 0B01, so that the image matrix sorting can be conveniently obtained by the number, thereby avoiding the need of sorting identification, and reducing the computational resource consumption.
And S3, the terminal self-evaluation module terminal performs self-evaluation on the terminal self-calculation capacity to obtain a terminal evaluation capacity B, for example, the mobile phone terminal can obtain the terminal evaluation capacity B by acquiring the model of the mobile phone and inquiring factory data, and can also adopt manual entry during installation.
S4, calculating a difference value between the evaluation value A and the reserved resource B to obtain an available resource C;
and S5, judging by the microscopic image processing module according to the customer grade, wherein the customer grade is a grade for collecting the business busy degree of the customer or the area where the customer is located, and the grade is set to be high when the business is busy or the distance between the business and the server is far, otherwise, the customer grade is low. When the customer level is high, the microscopic image processing module only performs microscopic image cropping processing.
And sending the cut microscopic image to a preprocessing server, preferentially processing incomplete microscopic image splicing and microscopic image identification processing of a terminal in a high bandwidth range, and then sending the microscopic image to a server side.
Example 3:
as shown in fig. 1 to 2, when the available resource C is greater than the computing resource required for the microscopic image clipping and the microscopic image splicing processing, but is less than the computing resource required for the microscopic image clipping, the microscopic image splicing and the microscopic image recognition processing, the microscopic image processing module performs the microscopic image clipping and the microscopic image splicing processing, the microscopic images are initially sorted according to the codes and then spliced, and the process of the microscopic image splicing comprises the following steps: visual field sub-block matching, visual field position fitting and block extraction;
as shown in fig. 3, the process of view sub-block matching is:
sa01, inputting, and initializing a result set M;
sa02, setting a current view i as a first view;
sa03, solving all adjacent view sets J of the current view i;
sa04, setting the current adjacent view J as the first view in J;
sa05, obtaining possible overlapping areas Ri and Rj of the visual field i and the visual field j;
sa06, rasterizing the template area Ri into a template sub-block set Pi;
sa07, arranging the template sub-block sets Pi in a descending order according to the dynamic range of the sub-blocks;
sa08, setting the current template sub-block P as the first one in the template sub-block set Pi;
sa09, finding a possible overlapping area s of the template subblocks P in the view J;
sa10, performing template matching search by taking the template sub-block P as a template and s as a search area;
sa11, adding the optimal matching M into a result set M;
sa12, finding all matching set visual field sets N consistent with M in the result set M;
sa13, comparing and judging whether the sum of the weights in the N is greater than a threshold value v;
if not, setting the current template sub-block P as the next template sub-block in the template sub-block set Pi, and returning to Sa09;
if yes, the next step is carried out;
sa14, comparing and judging whether the view J is the last view in the view set J;
if not, setting the view J as the next view in the view set J, and returning to the Sa05;
if yes, the next step is carried out;
sa15, comparing and judging, wherein the visual field i is the last visual field;
if not, setting i as the next visual field, and returning to Sa03;
and if so, outputting the result.
According to the scheme, the overlapping area of the microscopic images is identified, and the microscopic images are accurately sequenced.
As shown in fig. 4, the process of fitting the field of view position is:
sa16, inputting, and initializing all the view positions Xi and Yi;
sa17, setting the current view i as a first view;
sa18, obtaining a matching subset Mi containing the view i in the sub-block matching set M;
sa19, recalculating the positions Xi and Yi of the field of view i according to the matching subset Mi;
sa20, judging, and finishing updating all the visual fields;
if not, setting the view i as the next view;
if yes, the next step is carried out;
sa21, calculating the deviation average value L of the current wheel visual field position and the upper wheel visual field position;
sa22, comparing and judging, wherein the deviation average value L is smaller than a threshold value 1;
if not, returning to Sa17;
if yes, the next step is carried out;
sa23, visual field position normalization adjustment;
all the fields of view are output.
As shown in fig. 5, the process of block extraction is:
sa24, extracting the full graph size W, H;
sa25, dividing the whole image into a set B of blocks according to the block size;
sa26, calculating the positions of all blocks B in the set B;
sa27, setting block B as the first block in set B;
sa28, calculating a set Fb of all views overlapping block b;
sa29, setting the visual field f as the first visual field in Fb;
sa30, finding the overlapping areas Rb and Rf of the visual field f and the block b;
sa31, copy Rf in-image to Rb;
sa32, judging that the view f is the last view in the set Fb;
if not, setting the visual field f as the next visual field in the Fb, and returning to the Sa29;
if yes, the next step is carried out;
sa33, save the block b picture;
sa34, judging that the block B is the last block in the set B;
if not, setting the block B as the first block in the set B, and returning to Sa28;
and if so, outputting the result. According to the scheme, the positions are finely adjusted according to the overlapping areas among the sub-images, so that the cell positions are accurately spliced. And sending the spliced microscopic image to a server side, and finishing microscopic image identification operation by the server side.
When the microscopic image is spliced, whether the oblique diagonal pixel blocks are adjacent or not is processed, then a spline curve of the picture outline is fitted, and control points of the spline curve are optimized, so that the spliced microscopic image is smooth and jagged to the maximum extent.
Example 4:
when the available resource C is larger than the computing resources required by the microscopic image cutting, the microscopic image splicing and the microscopic image identification processing, the microscopic image processing module performs the microscopic image cutting, the microscopic image splicing and the microscopic image identification processing, displays the microscopic image identification result and sends the microscopic image identification result to the server, for example, the mononuclear cells in the microscopic image are identified, and the process is as follows:
sa100, detecting characteristic points of cell nuclei;
the image is reduced to a number of different scales, preferably: 0.3, 0.15, 0.08; respectively extracting feature points;
sa101, primary screening, and screening out too similar characteristic points according to characteristic point coordinates to reduce repeated cell extraction; by the steps, the identification efficiency is greatly improved.
In this example, if the distance of the feature point is not more than half the radius of the cell, and half the radius is greater than 32, it is considered too close that the distance is less than 32 pixels, otherwise it is considered too close that the distance is less than half the radius of the cell. I.e. cell.l 1distanceto (d.center) < math.min (cell.radius 0.5, 32).
Sa102, subdividing and dividing by using a color difference threshold;
converting the picture into an LAB format, and performing Otsu threshold segmentation on the weighted sum of the B channel and the A channel after phase inversion to obtain a cell nucleus mask picture; in the prior art, the screening is performed by using gray values, but the gray values are difficult to distinguish for some subtle positions because the gray values generally have only one channel and the numerical range is only 1-255. And the combined scheme of the channel B and the channel A is adopted, so that the numerical range can be greatly improved and the screening precision is improved due to the two channels.
The weight is 0.7 of the inverse phase of the channel B and 0.3 of the channel A;
sa103, image morphology calculation;
one or more combination of erosion and dilation operations; the corrosion calculation and the dilation calculation are, for example, the calculation method in chinese patent document CN 106875404A.
Sa104, fine screening, namely screening out non-cells with the nuclear ratio lower than 0.3, the nuclear radius higher than 150 pixels and the nuclear radius lower than 10 pixels according to the nuclear ratio parameter; the kernel proportion = the kernel area/detection feature point radius circle area subdivided by the color difference threshold. Thereby allowing identification of cells in the microscopic image.
Example 5:
and the doctor diagnoses the identified microscopic image, rechecks the microscopic image, submits the diagnosis opinion operation, renders the diagnosed data into PDF, JPG and WORD format files according to a corresponding report template, and sends the report to the client according to the association of registration information and digital sample information in the system.
The above-described embodiments are merely preferred technical solutions of the present invention, and should not be construed as limiting the present invention, and the embodiments and features in the embodiments in the present application may be arbitrarily combined with each other without conflict. The scope of the present invention is defined by the claims, and is intended to include equivalents of the features of the claims. I.e., equivalent alterations and modifications within the scope hereof, are also intended to be within the scope of this invention.
Claims (9)
1. A microscopic image rapid processing system comprises a terminal and a server, and is characterized in that: the terminal comprises a microscopic image scanning module, a terminal self-evaluation module, a microscopic image sorting module and a microscopic image processing module;
the microscopic image scanning module is used for acquiring microscopic images;
the terminal self-evaluation module is used for evaluating computing resources of the terminal;
the microscopic image sorting module is used for sorting the microscopic images;
the microscopic image processing module is used for carrying out different processing on the microscopic image according to the terminal computing resources evaluated by the terminal self-evaluation module;
the method comprises the following specific steps:
s1, when a terminal runs a microscopic image scanning module for the first time, taking the maximum value of computing resources required by the microscopic image scanning module as a reserved resource A, and setting the microscopic image scanning module as the highest priority;
s2, scanning by a microscopic image scanning module to obtain a plurality of microscopic images, and sequencing the plurality of microscopic images by a microscopic image sequencing module;
s3, the terminal self-evaluation module terminal performs self-evaluation on the self-calculation capacity of the terminal to obtain a terminal evaluation capacity B;
s4, calculating a difference value between the evaluation value A and the reserved resource B to obtain an available resource C;
s5, the microscopic image processing module judges according to the calculation capacity of the available resource C or the grade of a client to carry out terminal operation;
the method comprises the following steps:
sa1, when available resource C is smaller than computing power resource required by microscopic image splicing or the level of a client is high, the microscopic image processing module only performs microscopic image cutting processing, and the cut microscopic image is sent to a terminal workstation or a preprocessing server;
sa2, when the available resource C is larger than the computational resource required by the microscopic image cutting and splicing processing and is smaller than the computational resource required by the microscopic image cutting, the microscopic image splicing and the microscopic image identification processing, the microscopic image processing module performs the microscopic image cutting and the microscopic image splicing processing, and the spliced microscopic image is sent to the server side;
and Sa3, when the available resource C is larger than the computing resources required by the microscopic image cutting, the microscopic image splicing and the microscopic image identification processing, the microscopic image processing module performs the microscopic image cutting, the microscopic image splicing and the microscopic image identification processing, displays the microscopic image identification result and sends the microscopic image identification result to the server side.
2. A microscopic image rapid processing system according to claim 1, characterized in that: the microscopic image processing module comprises functions of microscopic image cutting, microscopic image splicing and microscopic image identification.
3. A microscopic image rapid processing system according to claim 1, characterized in that: in the step S2, the microscopic image sorting module sorts the microscopic images, namely reads the control commands, adds the control commands of the X axis and the Y axis into the serial numbers of the microscopic images, and obtains the array sorting of the microscopic images through the serial numbers.
4. A microscopic image rapid processing system according to claim 2, characterized in that: in step Sa1, the customer grade is a grade for collecting the busy degree of the customer or the area where the customer is located, and the grade is set to be high when the customer is busy or the distance from the customer to the server is long, otherwise, the grade is low.
5. A microscopic image rapid processing system according to claim 2, characterized in that: in the step Sa1, the microscopic image is cut into 128 × 128 to 1024 × 1024 pixels.
6. A microscopic image rapid processing system according to claim 2, characterized in that: and in the step Sa2, the microscopic image splicing is used for processing whether the adjacent relation of the oblique diagonal pixel blocks is processed, then a spline curve of the picture outline is fitted, and control points of the spline curve are optimized, so that the microscopic image after splicing is smooth and subjected to sawtooth removal to the maximum extent on the basis of ensuring accurate identification.
7. The microscopic image rapid processing system according to claim 4, characterized in that: a terminal workstation is set at a high customer level and is connected with a terminal through a high-speed local area network to process microscopic image splicing and microscopic image identification;
the terminal workstation is a multi-core server.
8. A microscopic image rapid processing system according to claim 1, characterized in that: the microscope images were stored in the format jpg, tif or png.
9. A microscopic image rapid processing system according to claim 1, characterized in that: before the microscopic image scanning module is operated in the step S1, whether a terminal workstation or a preprocessing server exists is judged, if yes, only microscopic image cutting processing is carried out, the cut microscopic image is uploaded to the terminal workstation or the preprocessing server to carry out microscopic image splicing and microscopic image identification processing, and if not, the microscopic image scanning module is operated to execute the subsequent steps.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010407145.9A CN111651268B (en) | 2020-05-14 | 2020-05-14 | Quick processing system for microscopic image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010407145.9A CN111651268B (en) | 2020-05-14 | 2020-05-14 | Quick processing system for microscopic image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111651268A CN111651268A (en) | 2020-09-11 |
| CN111651268B true CN111651268B (en) | 2023-01-31 |
Family
ID=72346663
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010407145.9A Active CN111651268B (en) | 2020-05-14 | 2020-05-14 | Quick processing system for microscopic image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111651268B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100433060C (en) * | 2006-10-25 | 2008-11-12 | 麦克奥迪实业集团有限公司 | Splice, storage, and browsing method for full-automatic microscopic image |
| CN106063272A (en) * | 2014-01-02 | 2016-10-26 | 世宗大学校产学协力团 | Method for encoding multi-view video and apparatus therefor and method for decoding multi-view video and apparatus therefor |
| CN106875404A (en) * | 2017-01-18 | 2017-06-20 | 宁波摩视光电科技有限公司 | The intelligent identification Method of epithelial cell in a kind of leukorrhea micro-image |
| CN105931187B (en) * | 2016-04-29 | 2019-06-14 | 北京小米移动软件有限公司 | Image processing method and device |
| CN105957008B (en) * | 2016-05-10 | 2019-08-16 | 厦门美图之家科技有限公司 | The real-time joining method of panoramic picture and system based on mobile terminal |
-
2020
- 2020-05-14 CN CN202010407145.9A patent/CN111651268B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100433060C (en) * | 2006-10-25 | 2008-11-12 | 麦克奥迪实业集团有限公司 | Splice, storage, and browsing method for full-automatic microscopic image |
| CN106063272A (en) * | 2014-01-02 | 2016-10-26 | 世宗大学校产学协力团 | Method for encoding multi-view video and apparatus therefor and method for decoding multi-view video and apparatus therefor |
| CN105931187B (en) * | 2016-04-29 | 2019-06-14 | 北京小米移动软件有限公司 | Image processing method and device |
| CN105957008B (en) * | 2016-05-10 | 2019-08-16 | 厦门美图之家科技有限公司 | The real-time joining method of panoramic picture and system based on mobile terminal |
| CN106875404A (en) * | 2017-01-18 | 2017-06-20 | 宁波摩视光电科技有限公司 | The intelligent identification Method of epithelial cell in a kind of leukorrhea micro-image |
Non-Patent Citations (1)
| Title |
|---|
| 图像分割与目标检测在农业场景中的应用研究;金沙沙;《湖州师范学院硕士学位论文》;20220518;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111651268A (en) | 2020-09-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11586028B2 (en) | Mobile phone-based miniature microscopic image acquisition device and image stitching and recognition methods | |
| CN110794569B (en) | Cell micro microscopic image acquisition device and image identification method | |
| CN110807757B (en) | Image quality evaluation method and device based on artificial intelligence and computer equipment | |
| CN113936132B (en) | Method and system for detecting water pollution of chemical plant based on computer vision | |
| CN110797097B (en) | Artificial intelligence cloud diagnosis platform | |
| CN116416244A (en) | Crack detection method and system based on deep learning | |
| CN114742211B (en) | Convolutional neural network deployment and optimization method facing microcontroller | |
| CN110807732B (en) | Panoramic stitching system and method for microscopic images | |
| CN111126162A (en) | Method, device and storage medium for identifying inflammatory cells in image | |
| CN111651268B (en) | Quick processing system for microscopic image | |
| CN115205250A (en) | Pathological image lesion segmentation method and system based on deep learning | |
| CN115272225A (en) | Strip steel surface defect detection method and system based on countermeasure learning network | |
| CN114092467A (en) | Scratch detection method and system based on lightweight convolutional neural network | |
| CN110766668B (en) | Cell detection and identification system and method | |
| CN110738658B (en) | Image quality evaluation method | |
| CN116721288A (en) | Helmet detection method and system based on YOLOv5 | |
| CN114897823B (en) | Cytological sample image quality control method, system and storage medium | |
| CN112200275B (en) | Artificial neural network quantification method and device | |
| CN113784147B (en) | Efficient video coding method and system based on convolutional neural network | |
| CN115496761A (en) | AE-based method and system for phase-splitting screening of low power lens and storage medium | |
| CN115690704A (en) | LG-CenterNet model-based complex road scene target detection method and device | |
| CN114627308A (en) | Extraction method and system of bone marrow cell morphological characteristics | |
| CN114693693A (en) | Fetal corpus callosum ultrasonic image segmentation method based on convolutional neural network | |
| CN117252879B (en) | Microscopic line scanning imaging system for semiconductor detection and control method thereof | |
| CN110824691B (en) | Image splicing method of cell micro-microscopic image acquisition device based on mobile phone |
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 |