CN112800929B - Bamboo shoot quantity and high growth rate online monitoring method based on deep learning - Google Patents
Bamboo shoot quantity and high growth rate online monitoring method based on deep learning Download PDFInfo
- Publication number
- CN112800929B CN112800929B CN202110095461.1A CN202110095461A CN112800929B CN 112800929 B CN112800929 B CN 112800929B CN 202110095461 A CN202110095461 A CN 202110095461A CN 112800929 B CN112800929 B CN 112800929B
- Authority
- CN
- China
- Prior art keywords
- bamboo
- frame
- bamboo shoot
- forest image
- area detection
- 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
- 235000017166 Bambusa arundinacea Nutrition 0.000 title claims abstract description 222
- 235000017491 Bambusa tulda Nutrition 0.000 title claims abstract description 222
- 241001330002 Bambuseae Species 0.000 title claims abstract description 222
- 235000015334 Phyllostachys viridis Nutrition 0.000 title claims abstract description 222
- 239000011425 bamboo Substances 0.000 title claims abstract description 222
- 238000000034 method Methods 0.000 title claims abstract description 18
- 238000012544 monitoring process Methods 0.000 title claims abstract description 12
- 238000013135 deep learning Methods 0.000 title claims abstract description 11
- 238000001514 detection method Methods 0.000 claims abstract description 90
- 238000012163 sequencing technique Methods 0.000 claims abstract description 9
- 238000012216 screening Methods 0.000 claims abstract description 8
- 230000001174 ascending effect Effects 0.000 claims description 6
- 238000013136 deep learning model Methods 0.000 claims description 5
- 238000010586 diagram Methods 0.000 description 6
- 238000012545 processing Methods 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 238000012360 testing method Methods 0.000 description 3
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 description 2
- 238000002372 labelling Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 241000282414 Homo sapiens Species 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 229910002092 carbon dioxide Inorganic materials 0.000 description 1
- 239000001569 carbon dioxide Substances 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 235000001497 healthy food Nutrition 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 239000003550 marker Substances 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/22—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
-
- 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
- G06V20/68—Food, e.g. fruit or vegetables
-
- 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
- G06V2201/07—Target detection
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Computational Linguistics (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a deep learning-based online monitoring method for the quantity and high growth rate of bamboo shoots, which comprises the following steps: 1, regularly acquiring a bamboo forest image containing bamboo shoots through a fixed network camera; 2, acquiring target area detection frames and vertex coordinates of the detection frames by adopting the trained bamboo shoot depth learning model, wherein the number of the detection frames represents the number of the bamboo shoots; 3, sequencing all the detection frames, and storing the sequenced corresponding coordinates into a basic database or a temporary database; 4, screening the bamboo shoots; and 5, calculating the high growth rate of the bamboo shoots. The method provided by the invention can monitor the quantity and high growth rate of the bamboo shoots in the bamboo forest in real time, thereby judging the growth conditions of the bamboo shoots. The method has strong practicability and accurate result, and is suitable for various aspects of bamboo forest management, especially cultivation of bamboo shoot forests and under-forest economy.
Description
Technical Field
The invention relates to the field of intelligent agriculture and forestry, in particular to an online monitoring method for the quantity and the high growth rate of bamboo shoots based on deep learning, which is suitable for various aspects of bamboo forest management, especially cultivation of bamboo shoots and forest economy.
Background
Bamboo has wide distribution in the global range, and is known as the second forest in the world. China is the world with the most bamboo resources. Bamboo not only plays an important role in absorbing carbon dioxide, preventing water and soil loss, maintaining biological diversity and the like, but also provides various processing materials for human beings, and bamboo shoots are green healthy food. The quantity and high growth state of bamboo shoots are closely related to the quality and yield of bamboo shoots, and therefore, in bamboo forest management, bamboo shoot forest cultivation and under-forest economy, the quantity and high growth state of bamboo shoots need to be known in time.
Due to the large density and complex environment of bamboo forests, the quantity and the high growth rate of the bamboo shoots are difficult to obtain efficiently and accurately. The traditional bamboo shoot quantity mainly depends on manual counting, and the high growth rate is calculated after manual contact type height measurement is carried out mainly through tools such as a tape measure and a marker post. The computer vision and image measurement are a feasible solution, the bamboo forest image is obtained regularly through a monitoring camera, the bamboo forest image is transmitted to a cloud server in a wireless mode, an image processing and analyzing algorithm is designed, and the number and the high growth rate of the bamboo shoots are obtained. The core of the common image processing and analyzing algorithm is image segmentation, and the quantity and the height can be further calculated only by segmenting bamboo shoots. Due to the fact that influences of various adverse factors such as background and illumination are large, characteristics such as colors, textures and shapes need to be comprehensively applied to a segmentation algorithm, complexity is high, and the effect is poor. A series of image processing operations are required subsequently, and errors are further introduced.
Disclosure of Invention
The invention aims to overcome the defects in the prior art, and provides an online monitoring method for the number and the high growth rate of the bamboo shoots based on deep learning, so that the online automatic detection of the number and the high growth rate of the bamboo shoots can be realized, the growth conditions of the bamboo shoots can be judged, and corresponding measures can be taken.
The invention adopts the following technical scheme for solving the technical problems:
the invention relates to a method for monitoring the quantity and the high growth rate of bamboo shoots on line based on deep learning, which is characterized by comprising the following steps:
step 1: defining the number of currently acquired image frames as n, and initializing n to 1;
step 2: acquiring the nth bamboo forest image containing bamboo shoots;
and step 3: using trainedCarrying out bamboo shoot detection on the obtained nth frame of bamboo forest image by a good bamboo shoot deep learning model to obtain m in the nth frame of bamboo forest imagenA bamboo shoot area detection frame and coordinate values of four vertexes of each bamboo shoot area detection frame, wherein the number m of the bamboo shoot area detection framesnRepresenting the number of the bamboo shoots in the detected n-th frame bamboo forest image;
and 4, step 4: for m in the n-th frame bamboo forest imagenSequencing the bamboo shoot area detection frames according to a sequencing principle;
and 5: judging whether n is 1, if so, storing the coordinates corresponding to the sorted bamboo shoot area detection frames into a basic database; otherwise, storing the coordinates corresponding to the sorted bamboo shoot area detection frames into a temporary database;
step 6: judgment mn=mn-1If yes, directly executing the step 7; otherwise, m is judgedn-mn-1If the number is more than 0, increasing m in the n-th frame bamboo forest imagen-mn-1Storing the bamboo shoot area detection frames into a basic database according to a sorting principle, and returning to the step 2 to continue detection if the bamboo shoot area detection frames are not sorted; when n is 1, let mn=mn-1;
And 7: screening bamboo shoots;
screening coordinate information of the bamboo shoot area detection frames in the temporary database and the basic database by using a formula (1), if the formula (1) is met, indicating that the corresponding bamboo shoot detection frame is valid, and otherwise, indicating that the corresponding bamboo shoot detection frame is invalid;
in the formula (1), xeAnd yeRespectively representing the maximum abscissa error and the maximum ordinate error; x is the number ofi1 n-1The x-axis coordinate and x of the top point of the lower left corner of the sorted ith bamboo shoot area detection frame in the n-1 th frame of bamboo forest imagei1 nShowing the lower left of the detection frame of the i-th bamboo shoot area after sorting in the n-th bamboo forest imageX-axis coordinate of angular point, yi1 n-1The y-axis coordinate and the y-axis coordinate of the top of the lower left corner of the sorted ith bamboo shoot area detection frame in the n-1 th frame of bamboo forest imagei1 nRepresenting the y-axis coordinate of the top point of the lower left corner of the sorted ith bamboo shoot area detection frame in the nth frame of bamboo forest image; i ═ 1, mn];
And 8: the height p of the ith bamboo shoot in the (n-1) th frame bamboo forest image is obtained by the formula (2)i n-1And a height p in the n-th frame of the bamboo forest imagei n;
In the formula (2), yi2 n-1The y-axis coordinate and the y-axis coordinate of the top of the lower right corner of the sorted ith bamboo shoot area detection frame in the n-1 th frame of bamboo forest imagei2 nThe y-axis coordinate of the top point of the lower right corner of the sorted ith bamboo shoot area detection frame in the nth frame of bamboo forest image is represented, and yi3 n-1The y-axis coordinate and the y-axis coordinate of the top right corner vertex of the i-th bamboo shoot area detection frame sequenced in the n-1 th frame of bamboo forest imagei3 nRepresenting the y-axis coordinate of the top right corner vertex of the sorted ith bamboo shoot area detection frame in the nth frame of bamboo forest image;
and step 9: calculating the high growth rate of the ith bamboo shoot in the nth bamboo forest image by using the formula (3)
In the formula (3), αi nRepresents the high growth rate, p, of the ith bamboo shoot in the n-th bamboo forest imagei n-1P represents the height of the ith bamboo shoot in the (n-1) th frame bamboo forest imagei nThe height of the ith bamboo shoot in the n-th frame bamboo forest image is shown.
The method for monitoring the quantity and the high growth rate of the bamboo shoots on line based on the deep learning is also characterized in that the ordering principle in the step 4 is as follows:
and performing ascending sorting on each bamboo shoot area detection frame by using the x-axis coordinate of the vertex at the lower left corner, and performing ascending sorting by using the y-axis coordinate of the vertex at the lower left corner if the x-axis coordinates are the same, so as to obtain the sorted bamboo shoot area detection frames.
Compared with the prior art, the invention has the beneficial effects that:
1. according to the method, the trained bamboo shoot deep learning model is combined with a unique target detection frame vertex coordinate sorting, screening and calculating method, so that the quantity and the high growth rate of the bamboo shoots can be simultaneously and accurately acquired, and the on-line automatic monitoring requirement is met.
2. The method is used for detecting the bamboo shoots based on the deep learning principle, and basically, new samples are continuously obtained subsequently along with the increase of application time, the detection precision has continuously improved space, and a target detection model can obtain a better training result.
Drawings
FIG. 1 is a general flow diagram of the process of the present invention;
FIG. 2 is a block diagram showing the actual detection of bamboo shoots by the method of the present invention;
FIG. 3 is a schematic view of a bamboo shoot detection box according to the method of the present invention;
FIG. 4 is a schematic diagram of a detection box of the bamboo shoot after growing up.
Detailed Description
In this embodiment, as shown in fig. 1, an online monitoring method for bamboo shoot number and high growth rate based on deep learning is performed according to the following steps:
step 1: acquiring bamboo shoot images in a real forest environment through a network camera, uploading the bamboo shoot images to an FTP (file transfer protocol) server in Aliyun, and downloading enough bamboo shoot images as an initial data set by a computer;
step 2: the initial data set is rotated by 90 degrees, 180 degrees and 270 degrees, the brightness is increased by 1.5 times, and is reduced by 0.1 time, and the like, so that a new data set enhanced by 6 times is obtained;
and step 3: labeling all bamboo shoots in the new data set by using image labeling software labellimg, and performing image labeling on the new data set according to the following steps of 8: 1: 1, dividing the ratio into a training set, a test set and a verification set;
and 4, step 4: selecting a target detection model training data set based on YOLOv4, testing the training result by using a test set and a verification set, and dynamically adjusting parameters such as learning rate, confidence coefficient and the like to obtain an optimal bamboo shoot deep learning model;
and 5: defining the number of image frames currently acquired from a server as n, and initializing n to 1;
step 6: acquiring the nth bamboo forest image containing bamboo shoots;
and 7: adopting a trained bamboo shoot deep learning model to carry out bamboo shoot detection on the obtained nth frame of bamboo forest image to obtain m in the nth frame of bamboo forest imagenCoordinate values of four vertexes of each of the bamboo shoot region detection frames. Wherein the number m of the bamboo shoot area detection framesnThe number of bamboo shoots in the detected n-th frame bamboo forest image is shown. As shown in the actual detection block diagram of bamboo shoots in FIG. 2, there are 19 detection blocks in the area of bamboo shoots, which means that 19 bamboo shoots are detected;
and 8: for m in the n-th frame bamboo forest imagenSequencing the bamboo shoot area detection frames according to a sequencing principle;
sorting principle: and (4) sorting each bamboo shoot area detection frame in an ascending order according to the x-axis coordinate of the vertex at the lower left corner of the bamboo shoot area detection frame, and if the x-axis coordinates are the same, sorting each bamboo shoot area detection frame in an ascending order according to the y-axis coordinate of the vertex at the lower left corner of the bamboo shoot area detection frame, so that the sorted bamboo shoot area detection frames are obtained. FIG. 3 is a schematic diagram of the bamboo shoot detection frames of the n-1 th frame bamboo forest image, wherein the sequence obtained by the four bamboo shoot region detection frames according to the sorting rule is A, B, C, D. Wherein the black triangular region represents bamboo shoot, (x)i1 n-1,yi1 n-1)、(xi2 n-1,yi2 n-1)、(xi3 n-1,yi3 n-1) Respectively representing the vertex coordinates of the lower left corner, the lower right corner and the upper right corner of the ith bamboo shoot area detection frame after sequencing in the image;
and step 9: judging whether n is 1, if so, storing the coordinates corresponding to the sorted bamboo shoot area detection frames into a basic database; otherwise, storing the coordinates corresponding to the sorted bamboo shoot area detection frames into a temporary database;
step 10: judgment mn=mn-1If yes, directly executing step 11; otherwise, m is judgedn-mn-1If the number is more than 0, increasing m in the n-th frame bamboo forest imagen-mn-1Storing the bamboo shoot area detection frames into a basic database according to a sorting principle, and returning to the step 6 to continue detection if the bamboo shoot area detection frames are not stored in the basic database; when n is 1, let mn=mn-1;
Step 11: screening bamboo shoots;
FIG. 4 is a schematic diagram of a bamboo shoot detection box in the n-th frame of the bamboo forest image. Wherein, the non-overlapping parts of the two triangles represent the parts corresponding to the bamboo shoot length in the n-1 th frame of bamboo forest image, (x)i1 n,yi1 n)、(xi2 n,yi2 n)、(xi3 n,yi3 n) Respectively representing the vertex coordinates of the lower left corner, the lower right corner and the upper right corner of the ith bamboo shoot area detection frame after sorting in the image. Generally, the coordinates of the vertex of the lower left corner of a strain of bamboo shoots do not have too large error, and the error of the coordinates of the vertex of the lower left corner before and after the growth of the bamboo shoots is considered as xeAnd yeThe case (1); screening coordinate information of the bamboo shoot area detection frames in the temporary database and the basic database by using a formula (1), if the formula (1) is met, indicating that the corresponding bamboo shoot detection frame is valid, and otherwise, indicating that the corresponding bamboo shoot detection frame is invalid;
in the formula (1), xeAnd yeRespectively representing the maximum abscissa error and the maximum ordinate error; x is the number ofi1 n-1The lower left corner of the detection frame of the ith bamboo shoot area sequenced in the n-1 th bamboo forest image is representedX-axis coordinate of vertex, xi1 nThe x-axis coordinate and y of the top point of the lower left corner of the sorted ith bamboo shoot area detection frame in the nth frame of bamboo forest image are representedi1 n-1The y-axis coordinate and the y-axis coordinate of the top of the lower left corner of the sorted ith bamboo shoot area detection frame in the n-1 th frame of bamboo forest imagei1 nRepresenting the y-axis coordinate of the top point of the lower left corner of the sorted ith bamboo shoot area detection frame in the nth frame of bamboo forest image; i ═ 1, mn];
Step 12: the height p of the ith bamboo shoot in the (n-1) th frame bamboo forest image is obtained by the formula (2)i n-1And a height p in the n-th frame of the bamboo forest imagei n;
In the formula (2), yi2 n-1The y-axis coordinate and the y-axis coordinate of the top of the lower right corner of the sorted ith bamboo shoot area detection frame in the n-1 th frame of bamboo forest imagei2 nThe y-axis coordinate of the top point of the lower right corner of the sorted ith bamboo shoot area detection frame in the nth frame of bamboo forest image is represented, and yi3 n-1The y-axis coordinate and the y-axis coordinate of the top right corner vertex of the i-th bamboo shoot area detection frame sequenced in the n-1 th frame of bamboo forest imagei3 nRepresenting the y-axis coordinate of the top right corner vertex of the sorted ith bamboo shoot area detection frame in the nth frame of bamboo forest image;
step 13: calculating the high growth rate alpha of the ith bamboo shoot in the nth frame bamboo forest image by using the formula (3)i n:
In the formula (3), αi nRepresents the high growth rate, p, of the ith bamboo shoot in the n-th bamboo forest imagei n-1P represents the height of the ith bamboo shoot in the (n-1) th frame bamboo forest imagei nThe height of the ith bamboo shoot in the n-th frame bamboo forest image is shown.
Claims (2)
1. A bamboo shoot quantity and high growth rate on-line monitoring method based on deep learning is characterized by comprising the following steps:
step 1: defining the number of currently acquired image frames as n, and initializing n to 1;
step 2: acquiring the nth bamboo forest image containing bamboo shoots;
and step 3: adopting a trained bamboo shoot deep learning model to carry out bamboo shoot detection on the obtained nth frame of bamboo forest image to obtain m in the nth frame of bamboo forest imagenA bamboo shoot region detection frame and coordinate values of four vertexes of each bamboo shoot region detection frame, wherein the number m of the bamboo shoot region detection framesnRepresenting the number of the bamboo shoots in the detected n-th frame bamboo forest image;
and 4, step 4: for m in the n-th frame bamboo forest imagenSequencing the bamboo shoot area detection frames according to a sequencing principle;
and 5: judging whether n is equal to 1, if yes, storing the coordinates corresponding to the sorted bamboo shoot area detection frames into a basic database; otherwise, storing the coordinates corresponding to the sorted bamboo shoot area detection frames into a temporary database;
step 6: judgment mn=mn-1If yes, directly executing the step 7; otherwise, m is judgedn-mn-1If the number is more than 0, increasing m in the n-th frame bamboo forest imagen-mn-1Storing the bamboo shoot area detection frames into a basic database according to a sorting principle, and returning to the step 2 to continue detection if the bamboo shoot area detection frames are not sorted; when n is 1, let mn=mn-1;
And 7: screening bamboo shoots;
screening coordinate information of the bamboo shoot area detection frames in the temporary database and the basic database by using a formula (1), if the formula (1) is met, indicating that the corresponding bamboo shoot detection frame is valid, and otherwise, indicating that the corresponding bamboo shoot detection frame is invalid;
in the formula (1), xeAnd yeRespectively representing the maximum abscissa error and the maximum ordinate error; x is the number ofi1 n-1X-axis coordinate, x, of the top of the lower left corner of the sorted ith bamboo shoot region detection box in the n-1 th bamboo forest imagei1 nThe x-axis coordinate and y of the top point of the lower left corner of the sorted ith bamboo shoot area detection frame in the nth frame of bamboo forest image are representedi1 n-1The y-axis coordinate and the y-axis coordinate of the top of the lower left corner of the sorted ith bamboo shoot area detection frame in the n-1 th frame of bamboo forest imagei1 nRepresenting the y-axis coordinate of the top point of the lower left corner of the sorted ith bamboo shoot area detection frame in the nth frame of bamboo forest image; i ═ 1, mn];
And 8: the height p of the ith bamboo shoot in the n-1 frame bamboo forest image is obtained by the formula (2)i n-1And a height p in the n-th frame of the bamboo forest imagei n;
In the formula (2), yi2 n-1The y-axis coordinate and the y-axis coordinate of the top of the lower right corner of the sorted ith bamboo shoot area detection frame in the n-1 th frame of bamboo forest imagei2 nThe y-axis coordinate of the top point of the lower right corner of the sorted ith bamboo shoot area detection frame in the nth frame of bamboo forest image is represented, and yi3 n-1Representing the y-axis coordinate of the top right corner vertex of the i-th bamboo shoot area detection frame after being sequenced in the n-1 th frame of bamboo forest image, yi3 nRepresenting the y-axis coordinate of the top right corner vertex of the sorted ith bamboo shoot area detection frame in the nth frame of bamboo forest image;
and step 9: calculating the high growth rate of the ith bamboo shoot in the nth frame bamboo forest image by using the formula (3)
In the formula (3), αi nShows the high growth rate, p, of the ith bamboo shoot in the n-th bamboo forest imagei n-1P represents the height of the ith bamboo shoot in the (n-1) th frame bamboo forest imagei nThe height of the ith bamboo shoot in the n-th frame bamboo forest image is shown.
2. The on-line monitoring method for the quantity and the high growth rate of the bamboo shoots based on the deep learning as claimed in claim 1, wherein the sequencing principle in the step 4 is as follows:
and performing ascending sorting on each bamboo shoot area detection frame by using the x-axis coordinate of the vertex at the lower left corner, and performing ascending sorting by using the y-axis coordinate of the vertex at the lower left corner if the x-axis coordinates are the same, so as to obtain the sorted bamboo shoot area detection frames.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110095461.1A CN112800929B (en) | 2021-01-25 | 2021-01-25 | Bamboo shoot quantity and high growth rate online monitoring method based on deep learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110095461.1A CN112800929B (en) | 2021-01-25 | 2021-01-25 | Bamboo shoot quantity and high growth rate online monitoring method based on deep learning |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112800929A CN112800929A (en) | 2021-05-14 |
CN112800929B true CN112800929B (en) | 2022-05-31 |
Family
ID=75811510
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110095461.1A Active CN112800929B (en) | 2021-01-25 | 2021-01-25 | Bamboo shoot quantity and high growth rate online monitoring method based on deep learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112800929B (en) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110472575A (en) * | 2019-08-15 | 2019-11-19 | 中国农业大学 | A kind of string tomato maturation detection method based on deep learning and computer vision |
CN110717903A (en) * | 2019-09-30 | 2020-01-21 | 天津大学 | Method for detecting crop diseases by using computer vision technology |
CN111582337A (en) * | 2020-04-24 | 2020-08-25 | 东南大学 | Strawberry malformation state detection method based on small sample fine-grained image analysis |
Family Cites Families (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10241097B2 (en) * | 2015-07-30 | 2019-03-26 | Ecoation Innovative Solutions Inc. | Multi-sensor platform for crop health monitoring |
CN105321342B (en) * | 2015-12-07 | 2017-11-21 | 北京航空航天大学 | A kind of intersection vehicles queue length detection method based on video of taking photo by plane |
CN109726665B (en) * | 2018-12-25 | 2022-10-25 | 中国科学院合肥物质科学研究院 | Agricultural pest detection method based on dynamic trajectory analysis |
US11120552B2 (en) * | 2019-02-27 | 2021-09-14 | International Business Machines Corporation | Crop grading via deep learning |
CN110348394A (en) * | 2019-07-15 | 2019-10-18 | 广东名阳信息科技有限公司 | A method of detection video static object |
CN110827269B (en) * | 2019-11-11 | 2024-03-05 | 深圳前海微众银行股份有限公司 | Crop growth change condition detection method, device, equipment and medium |
CN111260720A (en) * | 2020-01-13 | 2020-06-09 | 浙江大学 | Target height measuring system based on deep learning method |
CN111311630B (en) * | 2020-01-19 | 2023-05-23 | 上海智勘科技有限公司 | Method and system for intelligently counting cargo quantity through video in warehouse management |
CN111640087B (en) * | 2020-04-14 | 2023-07-14 | 中国测绘科学研究院 | SAR depth full convolution neural network-based image change detection method |
CN111652867A (en) * | 2020-06-01 | 2020-09-11 | 智农科技有限公司 | Heavy-environment-free plant growth method based on deep morphological belief network |
CN111862048B (en) * | 2020-07-22 | 2021-01-29 | 浙大城市学院 | Automatic fish posture and length analysis method based on key point detection and deep convolution neural network |
CN112052797A (en) * | 2020-09-07 | 2020-12-08 | 合肥科大立安安全技术有限责任公司 | MaskRCNN-based video fire identification method and system |
-
2021
- 2021-01-25 CN CN202110095461.1A patent/CN112800929B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110472575A (en) * | 2019-08-15 | 2019-11-19 | 中国农业大学 | A kind of string tomato maturation detection method based on deep learning and computer vision |
CN110717903A (en) * | 2019-09-30 | 2020-01-21 | 天津大学 | Method for detecting crop diseases by using computer vision technology |
CN111582337A (en) * | 2020-04-24 | 2020-08-25 | 东南大学 | Strawberry malformation state detection method based on small sample fine-grained image analysis |
Also Published As
Publication number | Publication date |
---|---|
CN112800929A (en) | 2021-05-14 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110378909B (en) | Single wood segmentation method for laser point cloud based on Faster R-CNN | |
CN110490874A (en) | Weaving cloth surface flaw detecting method based on YOLO neural network | |
CN108960135B (en) | Dense ship target accurate detection method based on high-resolution remote sensing image | |
CN113012150A (en) | Feature-fused high-density rice field unmanned aerial vehicle image rice ear counting method | |
CN111540006B (en) | Plant stomata intelligent detection and identification method and system based on deep migration learning | |
CN110853015A (en) | Aluminum profile defect detection method based on improved Faster-RCNN | |
CN108921057B (en) | Convolutional neural network-based prawn form measuring method, medium, terminal equipment and device | |
CN107657603A (en) | A kind of industrial appearance detecting method based on intelligent vision | |
CN110782467B (en) | Horse body ruler measuring method based on deep learning and image processing | |
CN111462058B (en) | Method for rapidly detecting effective rice ears | |
CN113435282B (en) | Unmanned aerial vehicle image ear recognition method based on deep learning | |
CN110969654A (en) | Corn high-throughput phenotype measurement method and device based on harvester and harvester | |
CN112634237A (en) | Long bamboo strip surface defect detection method and system based on YOLOv3 improved network | |
Li et al. | SPM-IS: An auto-algorithm to acquire a mature soybean phenotype based on instance segmentation | |
CN113780144A (en) | Crop plant number and stem width automatic extraction method based on 3D point cloud | |
CN112861666A (en) | Chicken flock counting method based on deep learning and application | |
CN110188662A (en) | A kind of AI intelligent identification Method of water meter number | |
CN112800929B (en) | Bamboo shoot quantity and high growth rate online monitoring method based on deep learning | |
CN111161227B (en) | Target positioning method and system based on deep neural network | |
CN117079125A (en) | Kiwi fruit pollination flower identification method based on improved YOLOv5 | |
CN116524344A (en) | Tomato string picking point detection method based on RGB-D information fusion | |
CN112862826B (en) | Normal sample nondestructive generation method for surface defect detection task | |
CN114842326A (en) | Calibration-free sandalwood plant seedling shortage positioning method | |
CN106503749B (en) | A kind of automatic grading method and its system of caviar | |
CN117854061B (en) | Tunnel face trace crack extraction method and system based on three-dimensional point cloud |
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 |