CN212255107U - Real-time detection device for grain quality - Google Patents
Real-time detection device for grain quality Download PDFInfo
- Publication number
- CN212255107U CN212255107U CN202022181286.1U CN202022181286U CN212255107U CN 212255107 U CN212255107 U CN 212255107U CN 202022181286 U CN202022181286 U CN 202022181286U CN 212255107 U CN212255107 U CN 212255107U
- Authority
- CN
- China
- Prior art keywords
- grain
- flow channel
- cereal
- plate
- real
- 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
Landscapes
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The utility model provides a real-time detection device for cereal material quality, which comprises a mobile cereal image capturing device, an embedded processor and an interactive display screen, wherein a cereal runner bottom plate in the mobile cereal image capturing device is fixed with an outlet of a cereal conveyor through a fixed iron wire, and the cereal runner bottom plate is wound with elastic bolts at two sides of a cereal collecting bin through an adjustable iron wire, so that the mobile cereal image capturing device is installed; in the real-time detection process, grains in the grain conveyor flow out, part of the grains are intercepted by the lower bottom plate of the grain flow channel, and the CCD industrial camera exposes the inside of the flow channel at a fixed period to capture images and transmits the images to the embedded processor. The utility model discloses can intercept cereal image in the flow fast, adapt to the cereal real-time detection of multiple application scene, different varieties.
Description
Technical Field
The utility model belongs to agricultural intelligence equips the field, concretely relates to cereal article matter real-time detection device.
Background
China is a big agricultural country, and the progress of agricultural mechanization has been greatly advanced, but the degree of agricultural machinery intellectualization is still low, and the main reason is that sensors in the agricultural background are few. The importance of agricultural smart sensors in the agricultural field is becoming more and more evident, and the demand for smart sensors in the agricultural field is also increasing. During the operation of the harvesting machine, the quality detection of the harvested grains is always a difficult problem, and how to obtain proper grain characteristics for subsequent quality detection becomes a first problem to be overcome.
Visual detection has become a popular detection means in recent years due to the characteristics of no contact, high accuracy, simple principle and the like, but is less used in the field of agricultural intelligent detection. At present, the visual processing method is mature, so that the key problem of grain quality detection can be solved as long as a proper grain picture can be obtained. But the picture that stews is adopted mostly in the acquisition of grain picture among the prior art to carrying out outward appearance detection to grain, inefficiency, step are loaded down with trivial details, and the sample randomness of gathering grain is not enough, can't satisfy the real-time quick detection demand of grain quality.
SUMMERY OF THE UTILITY MODEL
Exist not enoughly among the prior art, the utility model provides a cereal matter real-time detection device intercepts cereal image in the flow fast, realizes the real-time quick detection of cereal quality.
The utility model discloses a realize above-mentioned technical purpose through following technological means.
A real-time detection device for grain quality comprises a flowing grain image capturing device, an embedded processor and an interactive display screen, wherein the flowing grain image capturing device comprises a CCD industrial camera, the CCD industrial camera is in signal connection with the embedded processor, and the embedded processor is connected with the interactive display screen;
the CCD industrial camera is fixed inside the camera fixing support, the camera fixing support is fixed above the grain runner top plate horizontal plate, the grain runner top plate horizontal plate at the position where the camera fixing support is fixed is provided with a square opening, and a plurality of LED light sources are fixed on the inner side of the grain runner top plate horizontal plate close to the square opening; the vertical plates on two sides of the grain flow channel top plate are fixedly connected with the grain flow channel bottom plate, one end of the grain flow channel bottom plate is longer than the horizontal plate of the grain flow channel top plate, the end of the grain flow channel bottom plate is aligned with the outlet of the grain conveyor, the lower end of the horizontal plate of the grain flow channel top plate corresponding to the end is welded with a bent grain guide plate, the bending direction of the grain guide plate is far away from the outlet of the grain conveyor, and the bending angle of the grain guide plate is 45-65 degrees; the height from the bottom end of the grain guide plate to the grain flow channel bottom plate is 0.5-1.5 cm;
a camera lens is fixed below the CCD industrial camera;
the grain flow channel bottom plate is fixed with an outlet of the grain conveyor through a fixed iron wire and is also wound with elastic bolts at two sides of the grain collection bin through adjustable iron wires;
the installation angle alpha of the flowing grain image capturing device is between 20 and 90 degrees.
In the technical scheme, a lens adjusting window is arranged on one side face of the lower half part of the camera fixing support.
In the technical scheme, the length of one end of the grain flow channel bottom plate is 3-8cm longer than that of the horizontal plate of the grain flow channel top plate.
In the technical scheme, the grain guide plate is made of a stainless steel plate.
The utility model has the advantages that: the utility model discloses a cereal image collection system that flows can gather and intercept cereal image in the flow with the certain frequency, and convey and be used for image processing in the treater, and show the original drawing of gathering and testing result in real time, compare in gathering static cereal picture and promoted efficiency greatly, and the picture that gathers compares in static picture quality and does not descend, do not have the smear, the randomness of sampling has been promoted, handle to subsequent picture and do not increase work load, the real-time of detection has been guaranteed. The utility model discloses can be applied to in the cereal quality real-time detection scene, install the operation performance index that detects the harvester in grain tank screw feeder exit, install and advance grain department and detect cereal quality etc. at storage grain granary for the harvester when reaping.
Drawings
FIG. 1 is a side view of a mobile grain image capture device of the present invention;
FIG. 2 is a perspective view of the flowing grain image capturing device of the present invention;
fig. 3 is an installation schematic diagram of the flowing grain image capturing device of the present invention;
FIG. 4 is an overall flow chart of the real-time grain quality detection of the present invention;
fig. 5 is a diagram of grains collected by the flowing grain image capturing device of the present invention;
FIG. 6 is a flow chart of the image processing of the present invention;
fig. 7 is the utility model discloses parameter autotune trained XGBoost classification recognition model's acquisition flow chart.
Wherein: 1-CCD industrial camera, 2-camera lens, 3-LED light source, 4-fastening bolt, 5-fixing screw, 6-camera fixing support, 7-grain flow channel top plate, 8-grain flow channel bottom plate, 9-fixing bolt, 10-lens adjusting window, 11-square opening, 12-grain guide plate, 13-grain conveyor, 14-square groove, 15-fixing iron wire, 16-adjustable iron wire, 17-elastic bolt, 18-grain collecting bin, 19-device installation angle alpha, 20-grain and 21-lifting scraper.
Detailed Description
The invention will be further described with reference to the drawings and the following examples, but the scope of the invention is not limited thereto.
The utility model relates to a real-time detection device of cereal matter quality, including flowing cereal image capture device, embedded treater and interactive display screen, as shown in fig. 1, 2, flowing cereal image capture device includes CCD industry camera 1, camera lens 2, LED light source 3, fastening bolt 4, fixed screw 5, camera fixed bolster 6, cereal runner roof 7, cereal runner bottom plate 8, fixing bolt 9, camera lens adjusting window 10, square opening 11 and cereal guide plate 12; the CCD industrial camera 1 is in signal connection with an embedded processor, and the embedded processor is connected with an interactive display screen.
As shown in fig. 1 and 2, the CCD industrial camera 1 is fixed inside the camera fixing bracket 6 by the fixing screw 5, and the camera lens 2 is fixed below the CCD industrial camera 1; the grain flow channel top plate 7 is formed by fixedly connecting an upper horizontal plate and vertical plates at two sides, the camera fixing support 6 is fixed above the horizontal plate of the grain flow channel top plate 7 through a fixing bolt 9, and the fixing position of the camera fixing support 6 is positioned at a square opening 11 formed above the grain flow channel top plate 7; a lens adjusting window 10 is arranged on one side surface of the lower half part of the camera fixing support 6, and the lens adjusting window 10 is a square opening and is used for adjusting the focal length and the aperture of the camera lens 2; a plurality of LED light sources 3 are fixed on the inner side of the horizontal plate of the grain flow channel top plate 7 close to the square opening 11 through nanometer double faced adhesive tapes, and the LED light sources 3 are always on when the CCD industrial camera 1 works; the vertical plates on the two sides of the grain flow channel top plate 7 are fixedly connected with the grain flow channel bottom plate 8 through the fastening bolts 4, one end of the grain flow channel bottom plate 8 is aligned with the vertical plates, the length of the other end of the grain flow channel bottom plate 8 is 3-8cm (5 cm is preferred in the embodiment) larger than that of the horizontal plate of the grain flow channel top plate 7, and when the grain flow channel top plate is used, the longer end of the grain flow channel bottom plate 8 is aligned with the outlet of the grain conveyor 13; the grain guide plate 12 is welded at the lower end of the horizontal plate of the grain flow channel top plate 7 corresponding to the longer end of the grain flow channel bottom plate 8, the bending angle of the grain guide plate 12 towards the other end of the grain flow channel bottom plate 8 is 45-65 degrees (53 degrees are preferred in the embodiment), and the height from the bottom end of the grain guide plate 12 to the grain flow channel bottom plate 8 is 0.5-1.5cm (1 cm is preferred in the embodiment); in this embodiment, the size of the square opening 11 is 5cm × 5cm, the grain deflector 12 is made of a stainless steel plate, and the number of the fastening bolts 4 is preferably 4.
As shown in fig. 3, the schematic installation diagram of the image capturing device for flowing grains is that two sides of the outlet of the grain conveyor 13 are respectively provided with a square groove 14, the square groove 14 is wound on one end of a fixed iron wire 15, the other end of the fixed iron wire 15 is wound on a fastening bolt 4 at the longer end of the grain flow channel bottom plate 8, the other pair of fastening bolts 4 is wound on one end of an adjustable iron wire 16, the other end of the adjustable iron wire 16 is wound on elastic bolts 17 at two sides of the grain collecting bin 18, and the angle between the image capturing device for flowing grains and the horizontal plane is adjusted by adjusting the length of the adjustable iron wire 16 wound on the elastic bolts 17, so that the best image capturing effect is achieved. The grain flow rate is adjusted by the conveying speed of the grain conveyor 13Degree determination, speed V of the lifting blade 21 of the conveyor2The adjustable range is 10cm/s-60cm/s, the installation angle alpha 19 of the device is 20-90 degrees, and the proper grain conveying speed and installation angle are adjusted according to different application scenes, so that the collected pictures are distributed reasonably and are convenient for subsequent processing; if the grain flow rate is high, the tightening bolt 17 is tightened, so that the installation angle alpha 19 is reduced, and the too high grain flow rate is reduced; if the grain flow speed is slow, the loose-tight bolt 17 is loosened, so that the installation angle alpha 19 is enlarged, and the grain flow with slow flow speed is accelerated.
As shown in fig. 4, which is a general flow chart of real-time grain quality detection, firstly, grains 20 in a grain conveyor 13 flow out under the driving of an elevating scraper 21, part of the grains are intercepted by a grain flow channel lower bottom plate 8 and flow on the grain flow channel lower bottom plate 8, then the flow form is fixed by a guide plate 12 and flows to the position right below a square opening 11, a CCD industrial camera 1 exposes the inside of the flow channel at a fixed period to capture images and transmits the images to an embedded processor, the processor processes the images by using OPENCV and XGBoost algorithms, and displays the acquired original images and the processing results on an interactive display screen in real time, and then the grains flow out of the flow channel and slide into a grain collection bin 18; as shown in FIG. 5, for the utility model discloses the picture that the cereal image capture device gathered flows, and cereal distributes more evenly in the picture of gathering, piles up less, has made things convenient for subsequent processing greatly.
As shown in fig. 6, the specific process of the processor processing the image by using OPENCV and XGBoost algorithms is as follows:
s1, initializing the CCD industrial camera 1, acquiring the state of the CCD industrial camera 1, if the CCD industrial camera 1 is not normally connected, prompting no camera by an interactive display screen, if the connection is normal, setting the exposure mode of the CCD industrial camera 1 as soft trigger, namely, the CCD industrial camera 1 receives a trigger signal from an embedded processor to trigger, and then setting exposure time; because the utility model discloses the object that well CCD industrial camera 1 was shot is cereal in the motion, if the exposure time of setting for is improper can cause the blurred problem of smear of image, consequently need set for suitable exposure time t: namely, in the exposure time of the camera, the moving distance of the object is smaller than the precision of the camera, the camera cannot identify the moving distance of the object, and no smear is generated, namely the flowing speed v x exposure time t of the grain is less than or equal to the precision p of the camera (v x t is less than or equal to p); in the utility model, the movement speed v of the grain is not more than 100mm/s, i.e. 0.1mm/ms, the resolution of the camera is 200 ten thousand pixels (1600 × 1200), the visual field size is 60mm × 60mm, so the precision p of the camera is 60/1600 ═ 0.0375 mm/pixel; and t is less than or equal to 0.375ms according to t x v less than or equal to p, so that the smear can be avoided as long as the exposure time of the camera is set within 0.375 ms.
And S2, the CCD industrial camera 1 exposes and collects pictures according to the set exposure time, transmits the pictures to the embedded processor, and stores the received pictures locally.
S3, preprocessing the picture, denoising the image by using a function GaussianBlur of OPENCV, graying the image by using a BGR2GRAY function, binarizing the image by using THRESHOLD | BINARY to distinguish the foreground and the background of the image, using useful foreground information as white and useless background information as black, extracting connected domains by using a FindContours function, storing the connected domains in a container, traversing all the connected domains, extracting characteristic data of each connected domain and storing the characteristic data in a CSV file.
And S4, importing the extracted feature data into an XGboost classification recognition model with parameters automatically adjusted and trained, and recognizing impurities, broken and complete grains in each communication area by the XGboost.
And S5, counting the number of pixels of each category by using OPENCV, and converting the number of pixels into a proportion for acquiring the real-time quality of the grains.
And S6, saving the obtained result in a local text, displaying the result on a display screen, returning to S2, and carrying out the next exposure.
As shown in fig. 7, the process of obtaining the trained XGBoost classification recognition model with automatic parameter tuning includes the following steps:
s1, sample image acquisition: the method comprises the steps of acquiring 100 pictures by using a grain image acquisition device indoors, preprocessing the images by using OPENCV, binarizing, removing the background, reserving the foreground, and then extracting and storing connected regions in all the pictures.
S2, sample data preparation: and (3) extracting the connected regions according to the following steps of 7: 3, dividing the ratio into a training set and a test set; respectively creating a training set and a test set CSV data file and marking column labels, wherein the column labels are selected characteristic bases which can distinguish different types of grains, and are R, G, B, pixel number, external rectangle length-width ratio, curvature and manual judgment results, then extracting R, G, B, pixel number, external rectangle length-width ratio and curvature of each communication area, and storing the extracted results and the manual judgment results of the communication areas into one line of the CSV data file according to corresponding columns; both the test set and the training set contained data that was manually judged to be impurities, broken, and whole grain.
S3, improving the XGboost algorithm through a training set and a testing set: training a training set by using an XGboost algorithm, adjusting training parameters by combining with a parameter automatic tuning algorithm, improving the training precision so as to improve the XGboost algorithm, and then saving an improved XGboost model, namely the XGboost classification recognition model with the parameters automatically tuned and trained; and testing the test set by using the XGboost classification recognition model with the parameters automatically adjusted and trained, judging the precision (if the precision does not meet the requirement, training the training set again), and preventing the overfitting condition.
The main idea of the XGBoost algorithm is to perform a second-order taylor expansion on the target loss function loss according to the following formula:
the objective function is chosen to be loglos, so there are:
wherein, Obj(t)As an objective function, w (y)i，yi t-1+ft(xi) Is a loss function, ft(xi) As a regularization function, ω (f)t) For a structural function, C is a constant term,
for the first order partial derivative of each sample,
a second order partial derivative for each sample;
the training parameters that the XGBoost algorithm needs to adjust are as follows:
(1) n _ estimator: also called num _ boosting _ rounds, is the number of maximum trees generated, i.e. the maximum number of iterations.
(2) left _ rate: also called eta, the step size of each step iteration.
(3) gamma: when a node is split, the node is split only if the value of the loss function is reduced after the node is split; gamma specifies the minimum loss function degradation value required by node splitting; the larger the value of gamma, the more conservative the algorithm; because the larger the gamma value, the more the loss function drops to split the node.
(4) subsample: the parameter controls the proportion of random sampling for each tree, and over-fitting can be avoided by reducing the subsample value; however, if this value is set too small, it may result in under-fitting.
(5) colsample _ byte: the default value is 1, and is typically set to about 0.8, which is used to control the ratio of the number of columns per random sample.
(6) max _ depth: the default value is 6, numbers between 3 and 10 are commonly used, and max _ depth is the maximum depth of the tree and is used for controlling overfitting; the larger max _ depth, the more specific the model learning.
(7) max _ delta _ step: this parameter limits the maximum step size of the weight change per tree, which means no constraint if the value of this parameter is 0.
(8) lambda: also known as reg _ lambda, defaults to 0, a weighted L2 regularization term, which is used to control the regularization portion of XGBoost, is helpful in reducing overfitting.
(9) alpha: also known as reg _ alpha, defaults to 0, the L1 regularization term of weight.
The main idea of the parameter automatic tuning algorithm is as follows: the initial set is randomly generated according to uniform distribution, then a quasi-solution is selected from the current set, various selection methods which can adopt an evolutionary algorithm are selected, then a network probability model is established for the selected set, the new quasi-solution is obtained from the sampling of the model, finally, the solution obtained by sampling is added into the original set again, and the process is repeated until the termination condition is met. The main flow of the algorithm is as follows: randomly generating an initial population p (0) by setting t to 0, and selecting a candidate solution S (t) from p (t); constructing a grid B meeting the requirements under certain selection rules and limiting conditions; generating a new solution O (t) according to the joint distribution function of the network B; replacing the partial solution in p (t) with O (t) to form a new population p (t + 1); if the termination condition (training precision does not rise for 50 times continuously) is not met, reselecting S (t) and continuing.
The method comprises the steps that an automatic parameter tuning algorithm is combined with an XGboost algorithm, a training parameter interval to be selected is set according to the thought of the automatic parameter tuning algorithm, parameters are trained continuously in the process of training the XGboost algorithm by combining the thought of the automatic parameter tuning algorithm, the recognition result obtained by each training parameter is scored through a scoring function, and the optimal training parameter is obtained finally; the main process is as follows: setting t to 0, setting an initial combination P (0) of parameters, and selecting a quasi-solution S (t) from P (t); according to the formula:
wherein, XtFor the next sampling position, mut-1(x) As the mean value, σ, over the position input locationt-1(x) Is the standard deviation of the measured data to be measured,
is a weight parameter. For the next sampling position XtTaking into account the maximization of the function, a regression model of the Gaussian process is constructed, and the sum of the mean values at the input positions is predictedSelecting the input position with the maximum sum of the mean value and the standard deviation as a next sampling point; then generating a new solution O (t) according to the joint distribution function of the network B, and replacing P (t) with O (t) to form a new combination P (t + 1); if the termination condition is not met, reselecting S (t) and continuing.
The embodiment is a preferred embodiment of the present invention, but the present invention is not limited to the above embodiment, and any obvious improvement, replacement or modification which can be made by those skilled in the art without departing from the essence of the present invention belongs to the protection scope of the present invention.
Claims (4)
1. The real-time detection device for the grain quality is characterized by comprising a flowing grain image capturing device, an embedded processor and an interactive display screen, wherein the flowing grain image capturing device comprises a CCD industrial camera (1), the CCD industrial camera (1) is in signal connection with the embedded processor, and the embedded processor is connected with the interactive display screen;
the CCD industrial camera (1) is fixed inside a camera fixing support (6), the camera fixing support (6) is fixed above a horizontal plate of a grain flow channel top plate (7), a square opening (11) is formed in the horizontal plate of the grain flow channel top plate (7) at the position where the camera fixing support (6) is fixed, and a plurality of LED light sources (3) are fixed on the inner side of the horizontal plate of the grain flow channel top plate (7) close to the square opening (11); the vertical plates on two sides of the grain flow channel top plate (7) are fixedly connected with the grain flow channel bottom plate (8), one end of the grain flow channel bottom plate (8) is longer than the horizontal plate of the grain flow channel top plate (7), the end is aligned with the outlet of the grain conveyor (13), the lower end of the horizontal plate of the grain flow channel top plate (7) corresponding to the end is welded with a bent grain guide plate (12), the bending direction of the grain guide plate (12) is far away from the outlet of the grain conveyor (13), and the bending angle of the grain guide plate (12) is 45-65 degrees; the distance between the bottom end of the grain guide plate (12) and the grain flow channel bottom plate (8) is 0.5-1.5 cm;
a camera lens (2) is fixed below the CCD industrial camera (1);
the grain flow channel bottom plate (8) is fixed with an outlet of a grain conveyor (13) through a fixed iron wire (15), and the grain flow channel bottom plate (8) is wound with elastic bolts (17) on two sides of a grain collection bin (18) through an adjustable iron wire (16);
the installation angle alpha of the flowing grain image capturing device is between 20 and 90 degrees.
2. The real-time grain quality detection device according to claim 1, wherein a lens adjusting window (10) is formed on one side surface of the lower half part of the camera fixing support (6).
3. The real-time grain quality detection device according to claim 1, wherein the length of one end of the grain flow channel bottom plate (8) is 3-8cm longer than the length of the horizontal plate of the grain flow channel top plate (7).
4. The real-time grain quality detection device according to claim 1, wherein the grain guide plate (12) is made of stainless steel plate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202022181286.1U CN212255107U (en) | 2020-09-28 | 2020-09-28 | Real-time detection device for grain quality |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202022181286.1U CN212255107U (en) | 2020-09-28 | 2020-09-28 | Real-time detection device for grain quality |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN212255107U true CN212255107U (en) | 2020-12-29 |
Family
ID=73981411
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202022181286.1U Active CN212255107U (en) | 2020-09-28 | 2020-09-28 | Real-time detection device for grain quality |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN212255107U (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112067616A (en) * | 2020-09-28 | 2020-12-11 | 江苏大学 | Real-time detection device and method for grain quality |
-
2020
- 2020-09-28 CN CN202022181286.1U patent/CN212255107U/en active Active
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112067616A (en) * | 2020-09-28 | 2020-12-11 | 江苏大学 | Real-time detection device and method for grain quality |
| CN112067616B (en) * | 2020-09-28 | 2024-06-11 | 江苏大学 | Real-time detection device and method for grain quality |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112067616B (en) | Real-time detection device and method for grain quality | |
| CN109447945B (en) | Quick counting method for basic wheat seedlings based on machine vision and graphic processing | |
| CN110084241B (en) | Automatic ammeter reading method based on image recognition | |
| CN107316289A (en) | Crop field spike of rice dividing method based on deep learning and super-pixel segmentation | |
| CN110427922A (en) | One kind is based on machine vision and convolutional neural networks pest and disease damage identifying system and method | |
| CN114882468B (en) | Self-adaptive adjustment method for angle of sweeping brush of sweeper | |
| CN108416814B (en) | Method and system for quickly positioning and identifying pineapple head | |
| CN105868784A (en) | Disease and insect pest detection system based on SAE-SVM | |
| CN113361315B (en) | Banana string identification method based on background saturation compression and difference threshold segmentation fusion | |
| CN212255107U (en) | Real-time detection device for grain quality | |
| CN113706566B (en) | Edge detection-based perfuming and spraying performance detection method | |
| CN115131346B (en) | Fermentation tank processing procedure detection method and system based on artificial intelligence | |
| CN112990103A (en) | String mining secondary positioning method based on machine vision | |
| CN109726665A (en) | A kind of agricultural pests detection method based on dynamic trajectory analysis | |
| CN113516070A (en) | Pig counting method | |
| CN113222959A (en) | Fresh jujube wormhole detection method based on hyperspectral image convolutional neural network | |
| CN110428374A (en) | A kind of small size pest automatic testing method and system | |
| CN114742785A (en) | Hydraulic joint cleanliness control method based on image processing | |
| CN110765905B (en) | Method and device for measuring proportion of impurity contained in harvested grains of combine harvester | |
| CN113807132A (en) | Method and device for identifying irrigation state of plant growing area and storage medium | |
| CN108830834B (en) | Automatic extraction method for video defect information of cable climbing robot | |
| CN103366161B (en) | The statistical method of a kind of trialeurodes vaporariorum quantity and system | |
| CN106855942B (en) | A kind of video background extracting method based on weighted sample | |
| Germain et al. | Non destructive counting of wheatear with picture analysis | |
| CN115861768A (en) | Honeysuckle target detection and picking point positioning method based on improved YOLOv5 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| GR01 | Patent grant | ||
| GR01 | Patent grant |