CN116310263A - Pointer type aviation horizon instrument indication automatic reading implementation method - Google Patents
Pointer type aviation horizon instrument indication automatic reading implementation method Download PDFInfo
- Publication number
- CN116310263A CN116310263A CN202310262938.XA CN202310262938A CN116310263A CN 116310263 A CN116310263 A CN 116310263A CN 202310262938 A CN202310262938 A CN 202310262938A CN 116310263 A CN116310263 A CN 116310263A
- Authority
- CN
- China
- Prior art keywords
- horizon
- reading
- indication
- pointer
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/10—Image acquisition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- 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
- G06V10/757—Matching configurations of points or features
-
- 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
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- 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
- G06V2201/02—Recognising information on displays, dials, clocks
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to the technical field of airborne electronic equipment detection, in particular to a pointer type aviation horizon instrument indication automatic reading implementation method, which comprises the following specific steps: step (1) pitch angle indication reading: including static test readings and dynamic test readings; reading inclination angle indication in the step (2): including static test readings and dynamic test readings. According to the invention, the dial is detected through deep learning, and a key point detection algorithm based on the deep learning is adopted, so that the recognition speed and detection precision of key points are improved, and the real-time detection of the indication of the horizon is realized. The method for detecting the dynamic response capability of the horizon instrument is used for evaluating the uniformity and the stability of the motion of the dial and the pointer of the horizon instrument, and real-time monitoring of the motion states of the dial and the pointer of the horizon instrument is realized. A fixed scene video anti-shake algorithm is provided to improve the shake problem of camera pictures, and the algorithm has the advantages of stability, accuracy and rapidity.
Description
Technical Field
The invention relates to the technical field of airborne electronic equipment detection, in particular to a pointer type aviation horizon instrument indication automatic reading implementation method.
Background
In pointer type aviation horizon instrument inspection and maintenance, the reading or scale indication judgment of the horizon instrument is completed manually, maintenance personnel can record the data of various instruments through the position of a human eye interpretation pointer to reflect the working running state of a detection product, and the problems of low manual efficiency, low manual reading precision, poor stability, difficult data storage and data management, inconvenience in later data analysis and fault prediction and the like exist. With the development of machine vision, the method adopts an image recognition technology to automatically read the indication number of the horizon, and the core technology comprises two parts of dial area detection, pointer extraction and detection reading calculation, wherein the traditional method is to detect the dial area firstly, then extract pointer lines and then calculate the reading, and the detection efficiency is low.
Disclosure of Invention
In order to solve the technical problems, the invention provides a pointer type aviation horizon instrument indication automatic reading implementation method. The method combines deep learning with the traditional method, combines detection and identification, extracts the positions of key information such as pointers, digital scale marks and the like when detecting dial areas, and calculates the readings, so that the method has the advantages of high detection speed, strong robustness and high accuracy.
The technical problems to be solved by the invention are realized by adopting the following technical scheme:
the automatic reading implementation method for the pointer type aviation horizon instrument indication comprises the following specific steps:
step (1) pitch angle indication reading: including static test readings and dynamic test readings;
step (11) static test reading:
firstly, detecting the scale marks of the horizon based on the YOLOv3 algorithm, realizing the calculation and reading of the pitch angle indication of the horizon, and then, based on a horizon dynamic response capability detection method, realizing the detection of the uniformity and the stability of the motion of the horizon dial;
step (12) dynamic test reading:
firstly, enabling the horizon instrument to be in a vibration state, and then eliminating vibration influence based on a fixed scene video shake elimination rapid algorithm to realize calculation and reading of pitch angle indication of the horizon instrument;
reading inclination angle indication in the step (2): including static test readings and dynamic test readings;
step (21) static test reading:
firstly, detecting key points of the horizon based on deep learning to realize calculation and reading of inclination angle readings of the horizon, and then, based on a horizon dynamic response capability detection method, realizing uniformity and stability detection of pointer movement of the horizon;
step (22) dynamic test reading:
firstly, the horizon instrument is in a vibration state, and then vibration influence is eliminated based on a fixed scene video anti-shake fast algorithm to realize horizon instrument inclination angle indication calculation and reading.
Preferably, the horizon calibration line detection in the step (11) based on the YOLOv3 algorithm is specifically:
dividing the scale mark targets to be detected into 5 classes according to the structural characteristics of the horizon; labeling different kinds of targets through Labelme, and making a data set; model training is carried out through a YOLOv3 algorithm, and different targets are identified and extracted through the YOLOv3 algorithm.
Preferably, in the step (11), the calculation and reading of the pitch angle indication of the horizon is specifically:
firstly, the trained neural network is used for identifying and detecting the horizon graduation line, the world graduation line and the like of the horizon instrument, extracting the center of a circle of the identified key point, extracting the center point of the graduation line, and obtaining the pitch angle by calculating the distance between the center of the key point and the center point of the center line of different graduation lines.
Preferably, the method for detecting the dynamic response capability of the horizon in the step (11) and the step (21) specifically comprises the following steps:
aiming at uniformity and stability detection of movement of a dial and a pointer of the horizon, the uniformity and stability of movement of the dial and the pointer of the horizon are evaluated by continuously collecting multi-frame pictures, identifying scale marks and pointers of the pictures in a rapid target identification mode, extracting characteristic points, calculating relative movement speeds and tracks of the dial and the pointer between the multi-frame pictures according to the characteristic points.
Preferably, the fixed scene video jitter elimination fast algorithm in the step (12) and the step (22) is specifically:
aiming at camera picture jitter in detection, calculation precision and time cost are comprehensively considered, on the basis of taking a Harris corner detection operator and a pyramid LK optical flow method as motion estimation, the precision of jitter parameter estimation is improved by optimizing motion parameters through feature point layout, corner sub-pixelation, RANSAC eliminating foreground targets and a simplex method, and the quality of a reconstructed image in a jitter compensation link is improved by adopting an inverse mapping method and a bilinear interpolation algorithm.
Preferably, the horizon key point detection based on deep learning in step (21) is specifically:
aiming at two key points such as the pointer end point of the horizon and the digital scale, the key point detection algorithm in the deep learning technology is utilized for identification and extraction.
Preferably, the implementation of the horizon tilt angle indication calculation and reading in the step (21) is specifically:
clustering grouping is completed on all key points to obtain key point information of the horizon, and then the inclination angle is obtained by calculating the offset angle of the pointer relative to the key points of the digital scale on the basis.
The beneficial effects of the invention are as follows:
compared with the prior art, the invention has the remarkable advantages that:
1. the invention provides a method for combining deep learning with a traditional image processing algorithm, which detects a dial through the deep learning, adopts a key point detection algorithm based on the deep learning, improves the recognition speed and the detection precision of key points, and realizes the real-time detection of the indication number of a horizon.
2. The invention provides a method for detecting dynamic response capability of a horizon instrument, which is used for evaluating uniformity and stability of motion of a dial and a pointer of the horizon instrument, and realizing real-time monitoring of motion states of the dial and the pointer of the horizon instrument.
3. The invention provides a fixed scene video jitter elimination algorithm to improve the jitter problem of a camera picture, and the algorithm has the advantages of stability, accuracy and rapidity.
Drawings
The invention is further illustrated by the following examples in conjunction with the accompanying drawings:
FIG. 1 is a flow chart of a reading method of horizon indication according to the present invention;
fig. 2 is a flow chart of the debounce algorithm of the present invention.
Detailed Description
In order that the manner in which the invention is attained, as well as the features and advantages thereof, will be readily understood, a more particular description of the invention will be rendered by reference to the appended drawings.
An automatic reading implementation method for pointer type aviation horizon instrument readings comprises two aspects of pitch angle readings and inclination angle readings, as shown in fig. 1. The following is a detailed description of the operation procedure in combination with the reading process of the pitch angle and the inclination angle indication of the horizon:
step (1) pitch angle indication reading: including static test readings and dynamic test readings;
step (11) static test reading:
and firstly, detecting the scale mark of the horizon based on the YOLOv3 algorithm, and realizing the calculation and reading of the pitch angle indication of the horizon.
The method comprises the following steps: dividing the scale mark targets to be detected into 5 classes according to the structural characteristics of the horizon; labeling different kinds of targets through Labelme, and making a data set; model training is carried out through a YOLOv3 algorithm, and different targets are identified and extracted through the YOLOv3 algorithm.
And identifying and detecting the horizon graduation line, the world graduation line and the like of the horizon instrument by the trained neural network, extracting the center of a circle of the identified key point, extracting the center point of the graduation line, and obtaining a pitch angle by calculating the distances between the center of the key point and the center points of the center lines of different graduation lines.
And then based on the horizon dynamic response capability detection method, the uniformity and stability detection of the horizon dial movement are realized.
The method comprises the following steps: the method for detecting the dynamic response capability of the horizon comprises the following specific processes:
and acquiring color images of a multi-frame horizon panel through a camera, detecting and identifying the scale marks of the horizon instrument on the basis of a YOLOv3 algorithm for each frame of image, and extracting characteristic points.
After the pitch scale mark of the horizon is detected by the YOLOv3 algorithm, the center point is extracted, and the dynamic detection of the pitch angle change of the corresponding horizon can be realized by analyzing and calculating the magnitude and the speed of the position change of the center point within a certain time.
Step (12) dynamic test reading:
and the horizon instrument is in a vibration state, and then the influence of the vibration elimination of the video of the fixed scene is eliminated based on a quick algorithm for eliminating the vibration, so that the calculation and the reading of the pitch angle indication of the horizon instrument are realized.
According to the inspection process of the position indicator, the position indicator is required to be inspected under the vibration condition, and when vibration detection is carried out, the position of the position indicator and a camera for collecting images is relatively fixed, but vibration of a picture collected by the camera is inevitably caused. In order to ensure the stability of the images acquired by the camera, the scheme adopts a fast algorithm for eliminating the shake of the video of the fixed scene.
The algorithm flow of the fixed scene video anti-shake fast algorithm is shown in fig. 2, and the specific algorithm steps are as follows:
and (A) reading a gray value of a frame of YUV format image from a memory, and calculating the corner quantity representing the feature intensity by using a Harris operator.
And (B) setting a reference window, finding out the pixel with the largest corner value in the window, and selecting the pixel as an alternative characteristic point if the pixel is larger than a discrimination threshold value. And sequencing all the candidate points, and shielding surrounding candidate points in sequence from the beginning of the maximum corner quantity to obtain the feature points of layout optimization.
And (C) sub-pixelating the corner points, and improving the image coordinates of the extracted feature points to decimal numbers so as to further improve the matching precision.
And (D) reading the gray value of the next frame of image from the memory, and filtering out the characteristic points in the current frame tracking reference frame by using the pyramid LK optical flow to obtain matched characteristic point pairs.
Step (E) uses RANSAC separation motion. Firstly, 4 characteristic point pairs are randomly extracted, and roughly estimated motion parameters are obtained by utilizing a homography model. And measuring the matching condition of the matched feature points by using the parameters, wherein if the residual error of the motion estimation point and the matching point is larger than a reference threshold value, the matching condition is the foreground target or the feature point with the matching error, otherwise, the matching condition is the background feature point. The process is repeatedly executed for a plurality of times, the set of parameters with the most successful pairing is found, and the characteristic point pairs of the set of backgrounds are recorded.
And (F) optimizing the motion parameters by using a simplex method, and reducing the residual sum between the motion estimation point and the actual matching point as much as possible.
And (G) optimizing the pixel value after motion compensation by using bilinear interpolation to make the images uniform.
And (H) outputting a stable image frame, judging whether the next frame is successfully read in, if so, continuing to step (C) -step (H), otherwise, ending.
Reading inclination angle indication in the step (2): including static test readings and dynamic test readings;
step (21) static test reading:
the key point detection of the horizon is firstly carried out based on deep learning, so that the calculation and reading of the inclination angle indication of the horizon are realized.
The method comprises the following steps: aiming at two key points such as the pointer end point of the horizon and the digital scale, the key point detection algorithm in the deep learning technology is utilized for identification and extraction.
And clustering and grouping all the key points to obtain key point information of the horizon, and calculating the offset angle of the pointer relative to the key points of the digital scale on the basis to obtain the inclination angle.
And then based on the horizon dynamic response capability detection method, the uniformity and stability detection of the pointer movement of the horizon are realized.
The method comprises the following steps: the method for detecting the dynamic response capability of the horizon comprises the following specific processes:
and acquiring color images of the multi-frame horizon instrument panel through a camera, detecting key points based on deep learning for each frame of image, and extracting characteristic points.
And performing circle fitting by using the obtained pointer end key points of the multi-frame pictures of the horizon and combining the pointer rotation center points obtained by the key point detection algorithm to obtain the actual position pointed by the pointer end at a certain moment. The included angle between the connecting line of the center point of the horizon instrument and the key point of the tail end of the pointer at a certain moment and the connecting line of the key point of the zero scale mark is the pointer azimuth angle, and the dynamic detection of the inclination angle change of the instrument is realized by detecting the pointer azimuth angle.
Step (22) dynamic test reading:
and the horizon instrument is in a vibration state, and then vibration influence is eliminated based on a fixed scene video anti-shake fast algorithm to realize calculation and reading of the inclination angle indication of the horizon instrument.
According to the inspection process of the position indicator, the position indicator is required to be inspected under the vibration condition, and when vibration detection is carried out, the position of the position indicator and a camera for collecting images is relatively fixed, but vibration of a picture collected by the camera is inevitably caused. In order to ensure the stability of the images acquired by the camera, the scheme adopts a fast algorithm for eliminating the shake of the video of the fixed scene.
The algorithm flow of the fixed scene video anti-shake fast algorithm is shown in fig. 2, and the specific algorithm steps are as follows:
and (A) reading a gray value of a frame of YUV format image from a memory, and calculating the corner quantity representing the feature intensity by using a Harris operator.
And (B) setting a reference window, finding out the pixel with the largest corner value in the window, and selecting the pixel as an alternative characteristic point if the pixel is larger than a discrimination threshold value. And sequencing all the candidate points, and shielding surrounding candidate points in sequence from the beginning of the maximum corner quantity to obtain the feature points of layout optimization.
And (C) sub-pixelating the corner points, and improving the image coordinates of the extracted feature points to decimal numbers so as to further improve the matching precision.
And (D) reading the gray value of the next frame of image from the memory, and filtering out the characteristic points in the current frame tracking reference frame by using the pyramid LK optical flow to obtain matched characteristic point pairs.
Step (E) uses RANSAC separation motion. Firstly, 4 characteristic point pairs are randomly extracted, and roughly estimated motion parameters are obtained by utilizing a homography model. And measuring the matching condition of the matched feature points by using the parameters, wherein if the residual error of the motion estimation point and the matching point is larger than a reference threshold value, the matching condition is the foreground target or the feature point with the matching error, otherwise, the matching condition is the background feature point. The process is repeatedly executed for a plurality of times, the set of parameters with the most successful pairing is found, and the characteristic point pairs of the set of backgrounds are recorded.
And (F) optimizing the motion parameters by using a simplex method, and reducing the residual sum between the motion estimation point and the actual matching point as much as possible.
And (G) optimizing the pixel value after motion compensation by using bilinear interpolation to make the images uniform.
And (H) outputting a stable image frame, judging whether the next frame is successfully read in, if so, continuing the step ((C) -step (H), otherwise, ending.
The foregoing has shown and described the basic principles, principal features and advantages of the invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, and that the above embodiments and descriptions are merely illustrative of the principles of the present invention, and various changes and modifications may be made without departing from the spirit and scope of the invention, which is defined in the appended claims. The scope of the invention is defined by the appended claims and equivalents thereof.
Claims (10)
1. An automatic reading implementation method for indicating numbers of a pointer type aviation horizon is characterized in that: the method comprises the following specific steps:
step (1) pitch angle indication reading: including static test readings and dynamic test readings;
step (11) static test reading:
firstly, detecting the scale marks of the horizon based on the YOLOv3 algorithm, realizing the calculation and reading of the pitch angle indication of the horizon, and then, based on a horizon dynamic response capability detection method, realizing the detection of the uniformity and the stability of the motion of the horizon dial;
step (12) dynamic test reading:
firstly, enabling the horizon instrument to be in a vibration state, and then eliminating vibration influence based on a fixed scene video shake elimination rapid algorithm to realize calculation and reading of pitch angle indication of the horizon instrument;
reading inclination angle indication in the step (2): including static test readings and dynamic test readings;
step (21) static test reading:
firstly, detecting key points of the horizon based on deep learning to realize calculation and reading of inclination angle readings of the horizon, and then, based on a horizon dynamic response capability detection method, realizing uniformity and stability detection of pointer movement of the horizon;
step (22) dynamic test reading:
firstly, the horizon instrument is in a vibration state, and then vibration influence is eliminated based on a fixed scene video anti-shake fast algorithm to realize horizon instrument inclination angle indication calculation and reading.
2. The method for automatically reading the indication of the pointer type aviation horizon according to claim 1, which is characterized in that: in the step (11), the horizon rule detection based on the YOLOv3 algorithm is specifically as follows:
dividing the scale mark targets to be detected into 5 classes according to the structural characteristics of the horizon; labeling different kinds of targets through Labelme, and making a data set; model training is carried out through a YOLOv3 algorithm, and different targets are identified and extracted through the YOLOv3 algorithm.
3. The method for automatically reading the indication of the pointer type aviation horizon according to claim 1, which is characterized in that: the method for realizing the calculation and reading of the pitch angle indication of the horizon instrument in the step (11) comprises the following steps:
firstly, the trained neural network is used for identifying and detecting the horizon graduation line, the world graduation line and the like of the horizon instrument, extracting the center of a circle of the identified key point, and extracting the center point of the graduation line.
4. The method for automatically reading the indication of the pointer type aviation horizon according to claim 3, wherein the method comprises the following steps: in the step (11), the calculation and reading of the pitch angle indication of the horizon instrument are realized specifically as follows: and then calculating the distance between the center of the key point and the center point of the center line of the different graduation lines to obtain the pitch angle.
5. The method for automatically reading the indication of the pointer type aviation horizon according to claim 1, which is characterized in that: the method for detecting the dynamic response capability of the horizon in the step (11) and the step (21) specifically comprises the following steps:
aiming at the uniformity and stability detection of the movement of the dial and the pointer of the horizon, the scale mark and the pointer of the picture are identified by continuously collecting multi-frame pictures and adopting a rapid target identification mode, and then the characteristic points are extracted.
6. The method for automatically reading the indication of the pointer type aviation horizon according to claim 5, wherein the method comprises the following steps: the method for detecting the dynamic response capability of the horizon finder in the step (11) and the step (21) specifically further comprises the following steps: and calculating the movement speed and track of the relevant dial and the pointer between the multi-frame pictures according to the characteristic points, and evaluating the uniformity and stability of the movement of the horizon dial and the pointer.
7. The method for automatically reading the indication of the pointer type aviation horizon according to claim 1, which is characterized in that: the fixed scene video jitter elimination fast algorithm in the step (21) and the step (22) is specifically as follows:
aiming at camera picture jitter in detection, calculation precision and time cost are comprehensively considered, and on the basis of taking a Harris corner detection operator and a pyramid LK optical flow method as motion estimation, the precision of jitter parameter estimation is improved by optimizing feature point layout, corner sub-pixelation, RANSAC eliminating a foreground target and optimizing motion parameters by a simplex method.
8. The method for automatically reading the indication of the pointer type aviation horizon according to claim 7, wherein the method comprises the following steps: the fixed scene video jitter elimination fast algorithm in the step (21) and the step (22) specifically further comprises: and an inverse mapping method and a bilinear interpolation algorithm are adopted to improve the quality of the reconstructed image in the jitter compensation link.
9. The method for automatically reading the indication of the pointer type aviation horizon according to claim 1, which is characterized in that: in the step (21), the horizon key point detection based on deep learning specifically comprises the following steps:
aiming at two key points such as the pointer end point of the horizon and the digital scale, the key point detection algorithm in the deep learning technology is utilized for identification and extraction.
10. The method for automatically reading the indication of the pointer type aviation horizon according to claim 1, which is characterized in that: the reading of the inclination angle indication calculation of the horizon instrument in the step (21) is specifically as follows:
clustering grouping is completed on all key points to obtain key point information of the horizon, and then the inclination angle is obtained by calculating the offset angle of the pointer relative to the key points of the digital scale on the basis.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310262938.XA CN116310263A (en) | 2023-03-17 | 2023-03-17 | Pointer type aviation horizon instrument indication automatic reading implementation method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310262938.XA CN116310263A (en) | 2023-03-17 | 2023-03-17 | Pointer type aviation horizon instrument indication automatic reading implementation method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116310263A true CN116310263A (en) | 2023-06-23 |
Family
ID=86784792
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310262938.XA Pending CN116310263A (en) | 2023-03-17 | 2023-03-17 | Pointer type aviation horizon instrument indication automatic reading implementation method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116310263A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116758087A (en) * | 2023-08-22 | 2023-09-15 | 邦世科技(南京)有限公司 | Lumbar vertebra CT bone window side recess gap detection method and device |
-
2023
- 2023-03-17 CN CN202310262938.XA patent/CN116310263A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116758087A (en) * | 2023-08-22 | 2023-09-15 | 邦世科技(南京)有限公司 | Lumbar vertebra CT bone window side recess gap detection method and device |
| CN116758087B (en) * | 2023-08-22 | 2023-10-31 | 邦世科技(南京)有限公司 | Lumbar vertebra CT bone window side recess gap detection method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102521560B (en) | Instrument pointer image identification method of high-robustness rod | |
| CN111950330B (en) | Pointer instrument indication detection method based on target detection | |
| CN112818988B (en) | Automatic identification reading method and system for pointer instrument | |
| CN112257676B (en) | Pointer type instrument reading method and system and inspection robot | |
| CN110580723B (en) | Method for carrying out accurate positioning by utilizing deep learning and computer vision | |
| CN112149667B (en) | Automatic reading method of pointer instrument based on deep learning | |
| CN110991360B (en) | Robot inspection point position intelligent configuration method based on visual algorithm | |
| CN111368906B (en) | Pointer type oil level meter reading identification method based on deep learning | |
| CN103149087B (en) | Follow-up window and digital image-based non-contact real-time strain measurement method | |
| CN111476785B (en) | Night infrared light-reflecting water gauge detection method based on position recording | |
| CN113688817A (en) | Instrument identification method and system for automatic inspection | |
| CN113657339A (en) | Instrument pointer counting and reading method and medium based on machine vision | |
| CN108364306B (en) | Visual real-time detection method for high-speed periodic motion | |
| CN116310263A (en) | Pointer type aviation horizon instrument indication automatic reading implementation method | |
| CN115331151A (en) | Video speed measuring method and device, electronic equipment and storage medium | |
| CN115019294A (en) | Pointer instrument reading identification method and system | |
| CN114627461A (en) | Method and system for high-precision identification of water gauge data based on artificial intelligence | |
| CN109903308B (en) | Method and device for acquiring information | |
| CN117953048A (en) | Swivel bridge attitude monitoring system and method based on computer vision | |
| CN114120186A (en) | Cable detection and identification method and system based on Hololens | |
| CN113837312A (en) | Method and device for evaluating running state of zinc oxide arrester | |
| CN110580697A (en) | Video image processing method and system for measuring thickness of fetal nape transparency from ultrasonic video image | |
| Ueda et al. | A two-stage deep learning-based approach for automatic reading of analog meters | |
| CN113591705B (en) | Inspection robot instrument identification system and method and storage medium | |
| CN115311447A (en) | Pointer instrument indicating number identification method based on deep convolutional neural network |
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 |