CN110647813A - Human face real-time detection and identification method based on unmanned aerial vehicle aerial photography - Google Patents
Human face real-time detection and identification method based on unmanned aerial vehicle aerial photography Download PDFInfo
- Publication number
- CN110647813A CN110647813A CN201910771567.1A CN201910771567A CN110647813A CN 110647813 A CN110647813 A CN 110647813A CN 201910771567 A CN201910771567 A CN 201910771567A CN 110647813 A CN110647813 A CN 110647813A
- Authority
- CN
- China
- Prior art keywords
- data
- value
- method based
- real
- identification method
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 27
- 238000011897 real-time detection Methods 0.000 title abstract description 6
- 238000001514 detection method Methods 0.000 claims abstract description 34
- 238000012545 processing Methods 0.000 claims abstract description 21
- 238000011176 pooling Methods 0.000 claims abstract description 6
- 238000013507 mapping Methods 0.000 claims abstract description 5
- 238000007781 pre-processing Methods 0.000 claims abstract description 4
- 238000010606 normalization Methods 0.000 claims description 10
- 238000013528 artificial neural network Methods 0.000 claims description 4
- 230000005284 excitation Effects 0.000 claims description 4
- 230000006870 function Effects 0.000 claims description 4
- 230000005764 inhibitory process Effects 0.000 claims description 3
- 238000010304 firing Methods 0.000 claims description 2
- 239000011159 matrix material Substances 0.000 claims description 2
- 230000001629 suppression Effects 0.000 abstract description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/751—Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/172—Classification, e.g. identification
Abstract
The invention discloses a human face real-time detection and identification method based on unmanned aerial vehicle aerial photography, which comprises the following steps: loading a data processing model, acquiring data by a camera, and writing image data; preprocessing data; compressing the converted image data and storing the image data in a shared memory; the processor reads the image data; carrying out convolution processing on the data; the pooling layer performs dimensionality reduction on the data; carrying out nonlinear mapping on the data; carrying out classification processing on the weighted sum of the data; performing non-maximum suppression on the data; face matching is carried out on adjacent video frames, and a tracking template is detected and updated through a fixed frame number; calculating the correlation degree of the obtained target and the detected target to enhance the matching accuracy; superposing the face frames and counting the faces; and displaying the plug flow. The invention can realize real-time face detection and identification and greatly reduce the false detection rate.
Description
Technical Field
The invention relates to the technical field of unmanned aerial vehicle monitoring, in particular to a human face real-time detection and identification method based on unmanned aerial vehicle aerial photography.
Background
The traditional face recognition is mainly applied to recognition scenes based on fixed cameras, the face recognition effect is poor, the flexibility is poor, and once personnel move or the cameras move, face recognition data are inaccurate; in the relative movement of the camera and the personnel, the real-time performance of the face recognition is poor.
Disclosure of Invention
The invention aims to solve the problems and provides a human face real-time detection and identification method based on unmanned aerial vehicle aerial photography, which comprises the following steps:
s1: the detection module loads a data processing model and waits for receiving image data;
s2, the camera collects data and writes the data into the memory;
s3, preprocessing data: converting the format of image data, normalizing the image data, and converting the data into RGB data;
s4, compressing the converted image data and storing the image data in a shared memory;
s5, the detection module reads the image data and processes the data;
s6, carrying out convolution processing on the convolution layer neural network through convolution check data;
s7, the pooling layer performs dimensionality reduction on the data, and selects the maximum pixel value in the convolution kernel as a dimensionality reduced value;
s8, the excitation layer carries out nonlinear mapping on the output data of the convolution processing: let alpha be the slope coefficient, x be the abscissa value, and g (x) be the result of the function, then
g(x)＝max(αx，x)；
S9, classifying the extracted feature data by the full connection layer through weighted summation: y is Wx, wherein W is a weight vector and x is a feature vector;
s10, performing non-maximum value inhibition on the detection result, and removing repeated detection of the same target;
s11, reading each frame data in the video, and setting a reference value of the frame spacing frame number according to the scene; judging whether the current frame number is a multiple of a reference value; if yes, carrying out face detection on the next frame, and updating a tracking template; if not, matching with the face detected in the front so as to track the face;
s12, judging the correlation degree of the targets in the front frame and the rear frame by a normalized cross-correlation matching algorithm, if the correlation degree is within a certain threshold value range, determining the targets as the same face targets, otherwise, determining the targets as different face targets, and achieving the purpose of face tracking;
s13, superposing the face frames and counting the faces; and displayed by plug flow.
The invention has the beneficial effects that: based on the view angle of the unmanned aerial vehicle, the invention can realize mobile shooting, carry out real-time deep learning network, carry out network optimization and training aiming at the data format of the camera, further reduce the network operation amount, realize real-time face detection and recognition, and simultaneously greatly reduce the false detection rate.
Drawings
FIG. 1 is a flow chart of the present invention.
Detailed Description
The invention will be further described with reference to the accompanying drawings in which:
as shown in fig. 1, the invention relates to a human face real-time detection and identification method based on unmanned aerial vehicle aerial photography, which comprises the following steps:
s1: the detection module loads a data processing model and waits for receiving image data;
s2, the camera collects data and writes the data into the memory;
s3, preprocessing data: converting the format of image data, normalizing the image data, and converting the data into RGB data;
s4, compressing the converted image data and storing the image data in a shared memory;
s5, the detection module reads the image data and processes the data;
s6, carrying out convolution processing on the convolution layer neural network through convolution check data;
s7, the pooling layer performs dimensionality reduction on the data, and selects the maximum pixel value in the convolution kernel as a dimensionality reduced value;
s8, the excitation layer carries out nonlinear mapping on the output data of the convolution processing: let alpha be the slope coefficient, x be the abscissa value, and g (x) be the result of the function, then
g(x)＝max(αx，x)；
S9, classifying the extracted feature data by the full connection layer through weighted summation: y is Wx, wherein W is a weight vector and x is a feature vector;
s10, performing non-maximum value inhibition on the detection result, and removing repeated detection of the same target;
s11, reading each frame data in the video, and setting a reference value of the frame spacing frame number according to the scene; judging whether the current frame number is a multiple of a reference value; if yes, carrying out face detection on the next frame, and updating a tracking template; if not, matching with the face detected in the front so as to track the face;
s12, judging the correlation degree of the targets in the front frame and the rear frame by a normalized cross-correlation matching algorithm, if the correlation degree is within a certain threshold value range, determining the targets as the same face targets, otherwise, determining the targets as different face targets, and achieving the purpose of face tracking;
s13, superposing the face frames and counting the faces; and displayed by plug flow.
Further, the normalizing the data in S3 specifically includes:
and setting x as the value of the current pixel value normalization, x as the current pixel value, and max as the maximum value of all pixels in the current frame, then: x is the number of*＝log10(x)/log10(max)。
Further, the step S2 of converting the data format specifically includes: conversion of 1920 × 1080 × 2 uyvy data into 1920 × 1080 × 3 RGB data:
wherein: y is brightness, u and v denote chromaticity, R denotes a red channel, G denotes a green channel, and B denotes a blue channel.
Further, the convolution processing in S6 specifically includes:
where R (i, j) is the convolution value at the image (i, j), k is the convolution kernel of n × n, n is 2a, and f (i, j) is the pixel value at the image (i, j).
Further, the specific process of S7 is as follows:
y(u，v)＝max{f(u+i，v+j)，i∈(0，n)，j∈(0，n)}；
wherein: n denotes the size of the kernel and y (u, v) denotes the maximum pixel value in the neighborhood n x n in the image after max-firing.
Further, the specific process of S10 is as follows:
wherein, the interration is the intersection of two target frames, the Union is the Union of the two target frames, and if IoU exceeds a set threshold, the two target frames are the same target frame;
further, the performing face matching on adjacent frames in S11 specifically includes:
wherein: t is the matching template, S is the image matrix, and D is the matching result at (i, j).
Further, the normalization product correlation algorithm in S12 specifically processes:
wherein: e (S)i，j) The average gray scale value of the subgraph at (i, j), e (T) the average gray scale value of the template, M × N the matching template size, and T the matching template.
The specific process is as follows:
the detection module loads a data processing model and waits for receiving image data; the camera collects data and writes the data into the memory.
Pretreatment: the data that airborne camera gathered are fixed format, fixed resolution, and the data of gathering are the uyvy format, 1920x 1080's two-channel data, need carry out the preliminary treatment to data, including data format conversion and normalization, specifically include: and setting x as the value of the current pixel value normalization, x as the current pixel value, and max as the maximum value of all pixels in the current frame, then: x is the number of*＝log10(x)/log10(max)。
After normalization, 1920 × 1080 × 2 uyvy data was converted into 1920 × 1080 × 3 rgb data:
wherein: y is brightness, u and v denote chromaticity, R denotes a red channel, G denotes a green channel, and B denotes a blue channel.
To reduce the amount of computation and maintain the accuracy of target detection, a 1920x1080 resolution is compressed to 640 x 640: let Q11、Q12、Q21And Q22The pixel coordinate positions of adjacent pixel points in the image are respectively (x)1，y1)、(x1，y2)、(x2，y1) And (x)2，y2) P is the point to be estimated, R1Is Q11And Q21Approximate evaluation point of, R2Is Q12And Q22P is R1And R2Then:
through the normalization processing and bilinear interpolation calculation, the original data is processed into a format suitable for a face detection algorithm and stored in a shared memory, as shown in a flow chart shown in fig. 1.
And the processor reads the image data, loads the data processing model and processes the data.
Convolution processing: carrying out convolution processing on data through convolution check in the neural network: and setting x as the value of the current pixel value normalization, x as the current pixel value, and max as the maximum value of all pixels in the current frame, then: x is the number of*＝log10(x)/log10(max)。
Performing pooling treatment: the pooling layer performs dimensionality reduction on the data, a max-posing mode is selected for dimensionality reduction, the calculated amount can be reduced, and the maximum value of the data in the convolution kernel is selected as the value after dimensionality reduction: setting: n represents the size of the kernel, y (u, v) represents the maximum pixel value in the neighborhood n x n in the image (u, v) as a result after max-posing, then: y (u, v) ═ max { f (u + i, v + j), i ∈ (0, n), j ∈ (0, n) }.
Excitation: non-linear mapping of the convolution processed output: assuming α is the slope coefficient, x is the abscissa parameter, and g (x) is the result of the function, then: g (x) max (α x, x).
The full connection layer classifies the extracted feature data through weighted summation: and y is Wx, wherein W is a weight vector and x is a feature vector.
By the above-mentioned feature extraction of the data, and the target detection and classification, the same target may be detected many times in the result, and in order to eliminate the phenomenon, the result is subjected to non-maximum suppression, and the process is as follows:
the intersections is the intersection of two target frames, the Union is the Union of the two target frames, the larger the IoU is, the higher the overlapping degree is, if the overlapping degree exceeds a certain threshold value, the same target frame is considered, and the selection is preferred.
Reading frame data, judging whether the frame number is a multiple of a reference value according to the reference value of the scene set frame number, if so, performing face matching on adjacent frames, updating a tracking template, otherwise, performing face matching on the adjacent frames, and performing face tracking;
further judging the absolute error and the target obtained by the algorithm and the detected target by normalizing the algorithm correlation coefficientAnd the correlation degree greatly reduces the false detection rate. The specific process is as follows: setting: e (S)i，j) (ii) the average gray value of the subgraph at (i, j), e (T) the average gray value of the template, M × N the size of the matching template, and T the matching template, then:
based on the view angle of the unmanned aerial vehicle, the invention can realize mobile shooting, carry out real-time deep learning network, carry out network optimization and training aiming at the data format of the camera, further reduce the network operation amount, realize real-time face detection and recognition, and simultaneously greatly reduce the false detection rate.
The technical solution of the present invention is not limited to the limitations of the above specific embodiments, and all technical modifications made according to the technical solution of the present invention fall within the protection scope of the present invention.
Claims (8)
1. A real-time face detection and identification method based on unmanned aerial vehicle aerial photography is characterized by comprising the following steps:
s1: the detection module loads a data processing model and waits for receiving image data;
s2, the camera collects data and writes the data into the memory;
s3, preprocessing data: converting the format of image data, normalizing the image data, and converting the data into RGB data;
s4, compressing the converted image data and storing the image data in a shared memory;
s5, the detection module reads the image data and processes the data;
s6, carrying out convolution processing on the convolution layer neural network through convolution check data;
s7, the pooling layer performs dimensionality reduction on the data, and selects the maximum pixel value in the convolution kernel as a dimensionality reduced value;
s8, the excitation layer carries out nonlinear mapping on the output data of the convolution processing: let alpha be the slope coefficient, x be the abscissa value, and g (x) be the result of the function, then
g(x)＝max(αx，x)；
S9, classifying the extracted feature data by the full connection layer through weighted summation: y is Wx, wherein W is a weight vector and x is a feature vector;
s10, performing non-maximum value inhibition on the detection result, and removing repeated detection of the same target;
s11, reading each frame data in the video, and setting a reference value of the frame spacing frame number according to the scene; judging whether the current frame number is a multiple of a reference value; if yes, carrying out face detection on the next frame, and updating a tracking template; if not, matching with the face detected in the front so as to track the face;
s12, judging the correlation degree of the targets in the front frame and the rear frame by a normalized cross-correlation matching algorithm, if the correlation degree is within a certain threshold value range, determining the targets as the same face targets, otherwise, determining the targets as different face targets, and achieving the purpose of face tracking;
s13, superposing the face frames and counting the faces; and displayed by plug flow.
2. The real-time human face detection and identification method based on unmanned aerial vehicle aerial photography according to claim 1, wherein the normalization processing of data in S3 specifically comprises:
and setting x as the value of the current pixel value normalization, x as the current pixel value, and max as the maximum value of all pixels in the current frame, then: x is the number of*＝log10(x)/log10(max)。
3. The real-time human face detection and identification method based on unmanned aerial vehicle aerial photography according to claim 1, wherein the step S3 of converting the data format specifically comprises: converting the 1920 × 1080 × 2 UYVY-formatted data into 1920 × 1080 × 3 RGB-formatted data:
wherein: y is brightness, U and V denote chromaticities, R denotes a red channel, G denotes a green channel, and B denotes a blue channel.
4. The real-time human face detection and identification method based on unmanned aerial vehicle aerial photography according to claim 1, wherein the convolution processing in S6 specifically comprises:
where R (i, j) is the convolution value at the image (i, j), k is the convolution kernel of n × n, n is 2a, and f (i, j) is the pixel value at the image (i, j).
5. The real-time human face detection and identification method based on unmanned aerial vehicle aerial photography according to claim 1, wherein the S7 specific process is as follows:
y(u，v)＝max{f(u+i，v+j)，i∈(0，n)，j∈(0，n)}；
wherein: n denotes the size of the kernel and y (u, v) denotes the maximum pixel value in the neighborhood n x n in the image after max-firing.
6. The real-time human face detection and identification method based on unmanned aerial vehicle aerial photography according to claim 1, wherein the S10 specific process is as follows:
wherein, the interration is the intersection of two target frames, the Union is the Union of the two target frames, and if IoU exceeds the set threshold, the same target frame is obtained.
7. The real-time human face detection and identification method based on unmanned aerial vehicle aerial photography according to claim 1, wherein the performing of human face matching of adjacent frames in S11 specifically comprises:
wherein: t is the matching template, S is the image matrix, and D is the matching result at (i, j).
8. The real-time human face detection and identification method based on unmanned aerial vehicle aerial photography according to claim 1, wherein the normalization product correlation algorithm in the step S12 is used for processing the specific process:
wherein: e (S)i，j) The average gray scale value of the subgraph at (i, j), e (T) the average gray scale value of the template, M × N the matching template size, and T the matching template.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910771567.1A CN110647813A (en) | 2019-08-21 | 2019-08-21 | Human face real-time detection and identification method based on unmanned aerial vehicle aerial photography |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910771567.1A CN110647813A (en) | 2019-08-21 | 2019-08-21 | Human face real-time detection and identification method based on unmanned aerial vehicle aerial photography |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110647813A true CN110647813A (en) | 2020-01-03 |
Family
ID=69009705
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910771567.1A Pending CN110647813A (en) | 2019-08-21 | 2019-08-21 | Human face real-time detection and identification method based on unmanned aerial vehicle aerial photography |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110647813A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112229845A (en) * | 2020-10-12 | 2021-01-15 | 国网河南省电力公司濮阳供电公司 | Unmanned aerial vehicle high-precision winding tower intelligent inspection method based on visual navigation technology |
| CN112418117A (en) * | 2020-11-27 | 2021-02-26 | 北京工商大学 | Small target detection method based on unmanned aerial vehicle image |
| CN114299596A (en) * | 2022-03-09 | 2022-04-08 | 深圳联和智慧科技有限公司 | Smart city face recognition matching method and system and cloud platform |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107390722A (en) * | 2017-07-28 | 2017-11-24 | 上海瞬动科技有限公司合肥分公司 | A kind of unmanned plane during flying angle intelligent control method |
| CN107784288A (en) * | 2017-10-30 | 2018-03-09 | 华南理工大学 | A kind of iteration positioning formula method for detecting human face based on deep neural network |
| CN109063593A (en) * | 2018-07-13 | 2018-12-21 | 北京智芯原动科技有限公司 | A kind of face tracking method and device |
| CN110008919A (en) * | 2019-04-09 | 2019-07-12 | 南京工业大学 | The quadrotor drone face identification system of view-based access control model |
-
2019
- 2019-08-21 CN CN201910771567.1A patent/CN110647813A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107390722A (en) * | 2017-07-28 | 2017-11-24 | 上海瞬动科技有限公司合肥分公司 | A kind of unmanned plane during flying angle intelligent control method |
| CN107784288A (en) * | 2017-10-30 | 2018-03-09 | 华南理工大学 | A kind of iteration positioning formula method for detecting human face based on deep neural network |
| CN109063593A (en) * | 2018-07-13 | 2018-12-21 | 北京智芯原动科技有限公司 | A kind of face tracking method and device |
| CN110008919A (en) * | 2019-04-09 | 2019-07-12 | 南京工业大学 | The quadrotor drone face identification system of view-based access control model |
Non-Patent Citations (3)
| Title |
|---|
| 余丹: "基于卷积神经网络的图像识别方法研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
| 李楠: "基于多尺度梯度相角的人脸图像对称性检测与定位", 《小型微型计算机系统》 * |
| 沈沛意等: "《DAVINCI技术剖析及实战应用开发指南》", 30 April 2012 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112229845A (en) * | 2020-10-12 | 2021-01-15 | 国网河南省电力公司濮阳供电公司 | Unmanned aerial vehicle high-precision winding tower intelligent inspection method based on visual navigation technology |
| CN112418117A (en) * | 2020-11-27 | 2021-02-26 | 北京工商大学 | Small target detection method based on unmanned aerial vehicle image |
| CN112418117B (en) * | 2020-11-27 | 2023-05-12 | 北京工商大学 | Small target detection method based on unmanned aerial vehicle image |
| CN114299596A (en) * | 2022-03-09 | 2022-04-08 | 深圳联和智慧科技有限公司 | Smart city face recognition matching method and system and cloud platform |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11882357B2 (en) | Image display method and device | |
| CN107463949B (en) | Video action classification processing method and device | |
| CN109903331B (en) | Convolutional neural network target detection method based on RGB-D camera | |
| CN113065558A (en) | Lightweight small target detection method combined with attention mechanism | |
| CN110532970B (en) | Age and gender attribute analysis method, system, equipment and medium for 2D images of human faces | |
| CN107194317B (en) | Violent behavior detection method based on grid clustering analysis | |
| CN109740721B (en) | Wheat ear counting method and device | |
| CN110647813A (en) | Human face real-time detection and identification method based on unmanned aerial vehicle aerial photography | |
| WO2018082308A1 (en) | Image processing method and terminal | |
| CN111325051A (en) | Face recognition method and device based on face image ROI selection | |
| CN111738344A (en) | Rapid target detection method based on multi-scale fusion | |
| WO2022073282A1 (en) | Motion recognition method based on feature interactive learning, and terminal device | |
| CN112633185A (en) | Image processing method and device | |
| CN110991256A (en) | System and method for carrying out age estimation and/or gender identification based on face features | |
| CN113781421A (en) | Underwater-based target identification method, device and system | |
| CN111047618B (en) | Multi-scale-based non-reference screen content image quality evaluation method | |
| CN114463677A (en) | Safety helmet wearing detection method based on global attention | |
| WO2019228450A1 (en) | Image processing method, device, and equipment, and readable medium | |
| CN114581318A (en) | Low-illumination image enhancement method and system | |
| CN112101260B (en) | Method, device, equipment and storage medium for identifying safety belt of operator | |
| CN113435367A (en) | Social distance evaluation method and device and storage medium | |
| CN116778473A (en) | Improved YOLOV 5-based mushroom offline real-time identification method and system | |
| CN110766655A (en) | Hyperspectral image significance analysis method based on abundance | |
| CN115311520A (en) | Passion fruit maturity detection and positioning method based on visual identification | |
| CN107194406A (en) | A kind of panorama machine vision target identification method based on CS characteristic values |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200103 |
|
| RJ01 | Rejection of invention patent application after publication |