CN111476785A - Night infrared light-reflecting water gauge detection method based on position recording - Google Patents
Night infrared light-reflecting water gauge detection method based on position recording Download PDFInfo
- Publication number
- CN111476785A CN111476785A CN202010313038.XA CN202010313038A CN111476785A CN 111476785 A CN111476785 A CN 111476785A CN 202010313038 A CN202010313038 A CN 202010313038A CN 111476785 A CN111476785 A CN 111476785A
- Authority
- CN
- China
- Prior art keywords
- water gauge
- image
- night
- water
- scale
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01F—MEASURING VOLUME, VOLUME FLOW, MASS FLOW OR LIQUID LEVEL; METERING BY VOLUME
- G01F23/00—Indicating or measuring liquid level or level of fluent solid material, e.g. indicating in terms of volume or indicating by means of an alarm
- G01F23/04—Indicating or measuring liquid level or level of fluent solid material, e.g. indicating in terms of volume or indicating by means of an alarm by dip members, e.g. dip-sticks
-
- 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
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- 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
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10048—Infrared image
-
- 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
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/30—Assessment of water resources
Abstract
The invention provides a night infrared light-reflecting water gauge detection method based on position recording, which adopts an infrared camera to shoot a water gauge and uses a neural network to identify and read a water level value and comprises the following steps; step A1, setting a reset point of a camera, and shooting to obtain a template image for registration; the template image is used for setting the water gauge scale information of the station in the detection method; step A2, before detection, registering the camera with the template image; step A3, correcting the shot night water gauge image according to the registration result, and then carrying out image preprocessing on the night water gauge image; step A4, inputting the preprocessed night water gauge image into a neural network for recognition; a5, screening and correcting the semantic segmentation result in the A4; step A6, carrying out scale calculation on the semantic segmentation result of the corrected water gauge to obtain water level data in the image; the invention can realize the night water gauge image recognition of the water gauge detection station only by updating the software.
Description
Technical Field
The invention relates to the technical field of water level observation, in particular to a night infrared light-reflecting water gauge detection method based on position recording.
Background
In the water level observation scheme commonly used today, people often use the fluviograph to realize the long-range and automatic acquisition of water level, but the accuracy that the fluviograph measured receives the influence of various factors such as surface of water fluctuation easily, so also can combine other detection information to ensure the accuracy that the water gauge detected in practical application. The water level measurement by using the water gauge is a very intuitive and common mode. In recent years, remote monitoring cameras have become one of the standard configurations for hydrological data acquisition in hydrological detection units. Because of the monitoring camera technology is mature relatively, the maintenance cost is low, and some water level monitoring units can adopt the mode that the water level meter measurement combines the water gauge monitoring image to read the water level when carrying out water level measurement.
With the rapid development of artificial intelligence and image recognition technology, target recognition and image segmentation algorithms based on deep learning are becoming mature. This makes it possible to perform high-precision automatic water gauge detection and water level measurement based on the monitor image.
The existing water gauge image detection algorithms are generally divided into two types, one type is segmentation and identification by using the traditional graphics algorithm, for example, image semantic segmentation is performed by using a Canny operator; the second type is that a deep learning algorithm is used for recognizing the water gauge image, and the water level is determined according to the characteristics of the water gauge or the characteristics of the water surface.
However, the existing solutions all but require that the scale on the water gauge be visible, even in night scenes, the effect that the picture taken by the camera is close to the daytime is required. To achieve the effect, the camera is required to achieve the shooting effect of starlight [1] or black light level, clear color imaging can be performed at night, or extra light supplement equipment needs to be additionally arranged. However, none of these proposals suggests how to perform water gauge recognition in night scenes without additional fill lighting.
The invention aims to provide a recognition method for a night infrared reflective water gauge without additional visible light supplementary light based on daytime water gauge position recording, and the method can use an image semantic segmentation technology based on deep learning.
Disclosure of Invention
The invention provides a night infrared reflective water gauge detection method based on position recording, and night water gauge image recognition of a water gauge detection station can be realized only by updating software.
The invention adopts the following technical scheme.
A night infrared light-reflecting water gauge detection method based on position recording is used for identifying the water level value of a water gauge at night through machine vision, and the method adopts an infrared camera of a shooting device to shoot the water gauge with an infrared light-reflecting surface and uses a neural network to identify and read the water level value and comprises the following steps;
Step A1, setting a reset point of the camera, and shooting the water gauge at the reset point by using the camera under the bright light environment to acquire a template image for registration; the template image is used for setting the water gauge scale information of the station in the detection method;
Step A2, before entering night, shooting the water gauge at the reset point position by using a camera to obtain a target image for registration, adjusting the brightness, white balance and contrast of the target image by taking the previously obtained template image as reference, then carrying out registration calculation on the adjusted target image to obtain the position deviation value of the image shot at the reset point, and forming a registration result;
Step A3, in the water gauge detection operation, shooting a night water gauge image at the reset point position by using a camera, carrying out offset correction on the shot image according to the registration result, and preprocessing the image after the offset correction to obtain a night water gauge image ROI area;
Step A4, inputting the ROI area of the preprocessed night water gauge image into a neural network, and reading the ROI area of the night water gauge image by the neural network to obtain semantic segmentation results of all light reflecting areas containing the water gauge;
Step A5, screening and correcting the semantic segmentation result in the step A4 by using the recognition position record of the water gauge scale when the shooting equipment is in daytime to judge whether a water gauge exists in the night water gauge image, and excluding the recognition result of a non-water gauge area appearing in the image;
And step A6, carrying out scale calculation on the semantic segmentation result of the water gauge after screening and correction to obtain water level data in the image.
The method comprises the steps of performing water level value recognition by using a neural network, performing water scale semantic segmentation on a water scale image for the neural network, calculating water scale scales by using preset site scale information according to the result of the water scale semantic segmentation, and adjusting brightness, white balance and contrast of a template image in step A1, wherein the template image is a daytime image; setting site scale information corresponding to the template image by using the image during registration; the station scale information comprises an image ROI (region of interest), a water gauge elevation, three scale mark identifications and water gauge reading corresponding to the scale mark identifications.
In step a2, the last color picture before entering the night, which is taken by using the camera at the reset point, is used as a target image for image registration; then, adjusting the brightness, white balance and contrast of the template image and the target image, and then carrying out registration operation by using the processed images;
The ambient time, on which step a2 is based, is the time of the last moment of the day to ensure that the time at which registration is performed differs from the time at which identification is performed at night by no more than twelve hours.
And the registration operation adopts an image registration technology based on sift and flann, and the deviation value of the image shot by the camera at the reset point during the water gauge detection operation is acquired by superposing the processed target image to the processed template image during the registration.
In step A3 and step a4, image preprocessing is performed, that is, brightness, white balance and contrast are adjusted for the acquired night water gauge image, an ROI region in the image is captured to form a detection picture, the detection picture is adjusted to an input specification required by a neural network, during detection, the detection picture meeting the input specification is input into the neural network, and the neural network performs recognition to further determine a scale calculation target in the image.
In step a5, when the recognition position of the water gauge and the linear extension region in the direction of the water gauge are set as the nighttime recognition position region in the daytime of the neural network, and the semantic segmentation results outside the nighttime recognition position region are excluded when the semantic segmentation results obtained by reading the nighttime picture are screened and corrected.
Before the scale calculation in the step A6, under the condition that the neural network carries out semantic segmentation on the light reflecting part of the night water gauge picture, a light reflecting area outside the position record of the water gauge in the semantic segmentation result is identified and deleted through calculation to form a candidate area; when a plurality of water gauge targets exist in the candidate area, screening according to the size and the position relation of the water gauge targets to determine a calculation target; if the calculation target cannot be determined, returning to the alarm that the water gauge is lost, and if the calculation target can be determined, entering a scale calculation stage;
During scale calculation, the neural network determines the position of the water level in the pre-defined scale mark according to the semantic segmentation result of the water scale in the calculation target; and calculating the height of the pixel value of each centimeter scale of the water gauge in the detected picture according to the corresponding relation between the mark of the scale mark of the water gauge and the scale mark of the water gauge, which is calculated in advance, and calculating the water level in the detected picture by comparing the height of the pixel value of the scale mark of the water gauge in the detected picture with the height of the pixel value of the water level.
The image shot by the infrared camera is a black-and-white image, and the infrared camera automatically supplements light to an object in the image in the shooting process, so that the reflecting surface of the object can be in a reflecting effect in the image shot by the infrared camera.
The infrared reflecting surface comprises an enamel surface, an aluminum alloy surface, a stainless steel surface or a material surface with infrared reflecting capacity.
The invention has the advantages that:
1. The water gauge identification scheme only needs the camera to have an infrared function, dependence on additional light supplementing equipment can be avoided, and cost overhead of purchasing and deploying the light supplementing equipment is reduced.
2. Aiming at the problem of inaccurate identification caused by camera shake, the scheme adds the process of image registration in the identification process, so that the water gauge scale calculation result is more accurate.
3. The identification result of the night reflective water gauge is corrected according to the daytime position record, so that the result is more accurate.
Drawings
The invention is described in further detail below with reference to the following figures and detailed description:
FIG. 1 is a schematic view of the water gauge position at a water gauge detection station;
FIG. 2 is a schematic diagram of the ROI area of the image of the water gauge and the mark of the scale mark in the scale information of the station;
FIG. 3 is a schematic diagram of the generation of images required for the training of a neural network;
FIG. 4 is a schematic flow diagram of the present invention;
FIG. 5 is a comparison of the registration process before and after processing the captured image of the infrared camera;
FIG. 6 is a schematic diagram of screening and correcting night recognition results according to day recognition data to eliminate light spot interference;
In the figure: 1-a water gauge; 2-scale mark; 3-semantic segmentation results outside the nighttime identified location area (light spots on the nighttime water surface in the figure).
Detailed Description
As shown in fig. 1-6, the method for detecting a night infrared reflective water gauge based on position recording is used for identifying the water level value of the water gauge at night by machine vision, and is characterized in that: the method adopts an infrared camera of shooting equipment to shoot a water gauge 1 with an infrared reflecting surface, and uses a neural network to identify and read a water level value, and comprises the following steps;
Step A1, setting a reset point of the camera, and shooting the water gauge at the reset point by using the camera under the bright light environment to acquire a template image for registration; the template image is used for setting the water gauge scale information of the station in the detection method;
Step A2, before entering night, shooting the water gauge at the reset point position by using a camera to obtain a target image for registration, adjusting the brightness, white balance and contrast of the target image by taking the previously obtained template image as reference, then carrying out registration calculation on the adjusted target image to obtain the position deviation value of the image shot at the reset point, and forming a registration result;
Step A3, in the water gauge detection operation, shooting a night water gauge image at the reset point position by using a camera, carrying out offset correction on the shot image according to the registration result, and preprocessing the image after the offset correction to obtain a night water gauge image ROI area;
Step A4, inputting the ROI area of the preprocessed night water gauge image into a neural network, and reading the ROI area of the night water gauge image by the neural network to obtain semantic segmentation results of all light reflecting areas containing the water gauge;
Step A5, screening and correcting the semantic segmentation result in the step A4 by using the recognition position record of the water gauge scale when the shooting equipment is in daytime to judge whether a water gauge exists in the night water gauge image, and excluding the recognition result of a non-water gauge area appearing in the image;
And step A6, carrying out scale calculation on the semantic segmentation result of the water gauge after screening and correction to obtain water level data in the image.
The method comprises the steps of performing water level value recognition by using a neural network, performing water scale semantic segmentation on a water scale image for the neural network, calculating water scale scales by using preset site scale information according to the result of the water scale semantic segmentation, and adjusting brightness, white balance and contrast of a template image in step A1, wherein the template image is a daytime image; setting site scale information corresponding to the template image by using the image during registration; the station scale information comprises an image ROI (region of interest), a water gauge elevation, three scale mark identifications 2 and water gauge reading corresponding to the scale mark identifications.
In step a2, the last color picture before entering the night, which is taken by using the camera at the reset point, is used as a target image for image registration; then, adjusting the brightness, white balance and contrast of the template image and the target image, and then carrying out registration operation by using the processed images;
The ambient time, on which step a2 is based, is the time of the last moment of the day to ensure that the time at which registration is performed differs from the time at which identification is performed at night by no more than twelve hours.
And the registration operation adopts an image registration technology based on sift and flann, and the deviation value of the image shot by the camera at the reset point during the water gauge detection operation is acquired by superposing the processed target image to the processed template image during the registration.
In step A3 and step a4, image preprocessing is performed, that is, brightness, white balance and contrast are adjusted for the acquired night water gauge image, an ROI region in the image is captured to form a detection picture, the detection picture is adjusted to an input specification required by a neural network, during detection, the detection picture meeting the input specification is input into the neural network, and the neural network performs recognition to further determine a scale calculation target in the image.
In step a5, when the recognition position of the water gauge and the linear extension region in the direction of the water gauge in daytime by the neural network are taken as the nighttime recognition position region, and the semantic segmentation results 3 located outside the nighttime recognition position region are excluded when the semantic segmentation results obtained by recognizing and reading the nighttime picture are subjected to the screening correction.
Before the scale calculation in the step A6, under the condition that the neural network carries out semantic segmentation on the light reflecting part of the night water gauge picture, a light reflecting area outside the position record of the water gauge in the semantic segmentation result is identified and deleted through calculation to form a candidate area; when a plurality of water gauge targets exist in the candidate area, screening according to the size and the position relation of the water gauge targets to determine a calculation target; if the calculation target cannot be determined, returning to the alarm that the water gauge is lost, and if the calculation target can be determined, entering a scale calculation stage;
During scale calculation, the neural network determines the position of the water level in the pre-defined scale mark according to the semantic segmentation result of the water scale in the calculation target; and calculating the height of the pixel value of each centimeter scale of the water gauge in the detected picture according to the corresponding relation between the mark of the scale mark of the water gauge and the scale mark of the water gauge, which is calculated in advance, and calculating the water level in the detected picture by comparing the height of the pixel value of the scale mark of the water gauge in the detected picture with the height of the pixel value of the water level.
The image shot by the infrared camera is a black-and-white image, and the infrared camera automatically supplements light to an object in the image in the shooting process, so that the reflecting surface of the object can be in a reflecting effect in the image shot by the infrared camera.
The infrared reflecting surface comprises an enamel surface, an aluminum alloy surface, a stainless steel surface or a material surface with infrared reflecting capacity.
The neural network is used for image semantic segmentation, is based on Unet + Resnet34, and is pre-trained on an ImageNet data set;
The collection method of the images required by the training of the neural network comprises the following steps;
B1, collecting a large number of night water gauge images, and labeling the images to form a labeled sample;
Step B2, dividing the ROI area on the marked sample to intercept the area;
And step B3, carrying out image augmentation on the sample subjected to ROI region segmentation processing to form an image required by training of the neural network.
In this example, the water gauge elevation is the altitude or height from the water bottom corresponding to the position where the water gauge reading is 0.
In this example, the neural network carries the running software system for the server of the remote management center.
Example (b):
When the night water gauge image is identified, spot light or other luminous objects sometimes exist in a picture, and therefore identification errors can be caused. Therefore, the night result screening and correction can be carried out according to the day water gauge identification position record. As shown in fig. 6, two gray lines represent the left and right position information of the water gauge recorded after a certain station is identified in the daytime; in identifying nighttime, we first use neural networks to semantically segment the nighttime water gauge image, which causes the network to incorrectly identify the light spots in the picture as part of the water gauge as well.
Therefore, the recognition results outside the two gray line regions can be excluded according to the daytime position record information, and finally the screening correction result in the lower right corner in fig. 6 is obtained.
Claims (10)
1. A night infrared light-reflecting water gauge detection method based on position recording is used for identifying the water level value of a water gauge at night through machine vision and is characterized in that: the method adopts an infrared camera of shooting equipment to shoot a water gauge with an infrared reflecting surface, and carries out water level value recognition by a neural network, and comprises the following steps;
Step A1, setting a reset point of the camera, and shooting the water gauge at the reset point by using the camera under the bright light environment to acquire a template image for registration; the template image is used for setting the water gauge scale information of the station in the detection method;
Step A2, before entering night, shooting the water gauge at the reset point position by using a camera to obtain a target image for registration, adjusting the brightness, white balance and contrast of the target image by taking the previously obtained template image as reference, then carrying out registration calculation on the adjusted target image to obtain the position deviation value of the image shot at the reset point, and forming a registration result;
Step A3, in the water gauge detection operation, shooting a night water gauge image at the reset point position by using a camera, carrying out offset correction on the shot image according to the registration result, and preprocessing the image after the offset correction to obtain a night water gauge image ROI area;
Step A4, inputting the ROI area of the preprocessed night water gauge image into a neural network, and reading the ROI area of the night water gauge image by the neural network to obtain semantic segmentation results of all light reflecting areas containing the water gauge;
Step A5, screening and correcting the semantic segmentation result in the step A4 by using the recognition position record of the water gauge scale when the shooting equipment is in daytime to judge whether a water gauge exists in the night water gauge image, and excluding the recognition result of a non-water gauge area appearing in the image;
And step A6, carrying out scale calculation on the semantic segmentation result of the water gauge after screening and correction to obtain water level data in the image.
2. The method for detecting the night infrared reflective water gauge based on the position record according to claim 1, characterized in that: the method comprises the steps of performing water level value recognition by using a neural network, performing water scale semantic segmentation on a water scale image for the neural network, calculating water scale scales by using preset site scale information according to the result of the water scale semantic segmentation, and adjusting brightness, white balance and contrast of a template image in step A1, wherein the template image is a daytime image; setting site scale information corresponding to the template image by using the image during registration; the station scale information comprises an image ROI (region of interest), a water gauge elevation, three scale mark identifications and water gauge reading corresponding to the scale mark identifications.
3. The method for detecting the night infrared reflective water gauge based on the position record according to claim 1, characterized in that: in step a2, the last color picture before entering the night, which is taken by using the camera at the reset point, is used as a target image for image registration; then, adjusting the brightness, white balance and contrast of the template image and the target image, and then carrying out registration operation by using the processed images;
The ambient time, on which step a2 is based, is the time of the last moment of the day to ensure that the time at which registration is performed differs from the time at which identification is performed at night by no more than twelve hours.
4. The method for detecting the night infrared reflective water gauge based on the position record according to claim 3, characterized in that: and the registration operation adopts an image registration technology based on sift and flann, and the deviation value of the image shot by the camera at the reset point during the water gauge detection operation is acquired by superposing the processed target image to the processed template image during the registration.
5. The method for detecting the night infrared reflective water gauge based on the position record according to claim 1, characterized in that: in step A3 and step a4, image preprocessing is performed, that is, brightness, white balance and contrast are adjusted for the acquired night water gauge image, an ROI region in the image is captured to form a detection picture, the detection picture is adjusted to an input specification required by a neural network, during detection, the detection picture meeting the input specification is input into the neural network, and the neural network performs recognition to further determine a scale calculation target in the image.
6. The method for detecting the night infrared reflective water gauge based on the position record according to claim 5, characterized in that: in step a5, when the recognition position of the water gauge and the linear extension region in the direction of the water gauge are set as the nighttime recognition position region in the daytime of the neural network, and the semantic segmentation results outside the nighttime recognition position region are excluded when the semantic segmentation results obtained by reading the nighttime picture are screened and corrected.
7. The method for detecting the night infrared reflective water gauge based on the position record according to claim 6, wherein the method comprises the following steps: before the scale calculation in the step A6, under the condition that the neural network carries out semantic segmentation on the light reflecting part of the night water gauge picture, a light reflecting area outside the position record of the water gauge in the semantic segmentation result is identified and deleted through calculation to form a candidate area; when a plurality of water gauge targets exist in the candidate area, screening according to the size and the position relation of the water gauge targets to determine a calculation target; if the calculation target cannot be determined, returning to the alarm that the water gauge is lost, and if the calculation target can be determined, entering a scale calculation stage;
During scale calculation, the neural network determines the position of the water level in the pre-defined scale mark according to the semantic segmentation result of the water scale in the calculation target; and calculating the height of the pixel value of each centimeter scale of the water gauge in the detected picture according to the corresponding relation between the mark of the scale mark of the water gauge and the scale mark of the water gauge, which is calculated in advance, and calculating the water level in the detected picture by comparing the height of the pixel value of the scale mark of the water gauge in the detected picture with the height of the pixel value of the water level.
8. The method for detecting the night infrared reflective water gauge based on the position record according to claim 1, characterized in that: the image shot by the infrared camera is a black-and-white image, and the infrared camera automatically supplements light to an object in the image in the shooting process, so that the reflecting surface of the object can be in a reflecting effect in the image shot by the infrared camera.
9. The method for detecting the night infrared reflective water gauge based on the position record according to claim 1, characterized in that: the infrared reflecting surface comprises an enamel surface, an aluminum alloy surface, a stainless steel surface or a material surface with infrared reflecting capacity.
10. The method for detecting the night infrared reflective water gauge based on the position record according to claim 1, characterized in that: the neural network is used for image semantic segmentation, is based on Unet + Resnet34, and is pre-trained on an ImageNet data set;
The collection method of the images required by the training of the neural network comprises the following steps;
B1, collecting a large number of night water gauge images, and labeling the images to form a labeled sample;
Step B2, dividing the ROI area on the marked sample to intercept the area;
And step B3, carrying out image augmentation on the sample subjected to ROI region segmentation processing to form an image required by training of the neural network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010313038.XA CN111476785B (en) | 2020-04-20 | 2020-04-20 | Night infrared light-reflecting water gauge detection method based on position recording |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010313038.XA CN111476785B (en) | 2020-04-20 | 2020-04-20 | Night infrared light-reflecting water gauge detection method based on position recording |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111476785A true CN111476785A (en) | 2020-07-31 |
| CN111476785B CN111476785B (en) | 2023-04-07 |
Family
ID=71754292
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010313038.XA Active CN111476785B (en) | 2020-04-20 | 2020-04-20 | Night infrared light-reflecting water gauge detection method based on position recording |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111476785B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113141469A (en) * | 2021-04-21 | 2021-07-20 | 东莞市水利勘测设计院有限公司 | Anti-reflection method, system and device for night observation water gauge and readable storage medium |
| CN113221898A (en) * | 2021-04-16 | 2021-08-06 | 北京科技大学 | Automatic water level gauge reading method |
| CN113657275A (en) * | 2021-08-16 | 2021-11-16 | 中国科学院空天信息创新研究院 | Automatic detection method for forest and grass fire points |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006185410A (en) * | 2004-12-02 | 2006-07-13 | Nissan Motor Co Ltd | Nighttime mobile object notification device and nighttime mobile object notification method |
| CN106557764A (en) * | 2016-11-02 | 2017-04-05 | 江西理工大学 | A kind of water level recognition methodss based on binary-coded character water gauge and image procossing |
| CN107367310A (en) * | 2017-07-11 | 2017-11-21 | 华南理工大学 | A kind of river level remote monitoring method based on computer vision |
| CN108073905A (en) * | 2017-12-21 | 2018-05-25 | 北京邮电大学 | A kind of method, system and the equipment of intelligence water gauge reading |
| CN109764930A (en) * | 2018-12-27 | 2019-05-17 | 河海大学 | A kind of water gauge water level line visible detection method suitable for complex illumination condition |
| CN110956172A (en) * | 2019-11-18 | 2020-04-03 | 四创科技有限公司 | Water gauge identification method based on image semantic segmentation |
-
2020
- 2020-04-20 CN CN202010313038.XA patent/CN111476785B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006185410A (en) * | 2004-12-02 | 2006-07-13 | Nissan Motor Co Ltd | Nighttime mobile object notification device and nighttime mobile object notification method |
| CN106557764A (en) * | 2016-11-02 | 2017-04-05 | 江西理工大学 | A kind of water level recognition methodss based on binary-coded character water gauge and image procossing |
| CN107367310A (en) * | 2017-07-11 | 2017-11-21 | 华南理工大学 | A kind of river level remote monitoring method based on computer vision |
| CN108073905A (en) * | 2017-12-21 | 2018-05-25 | 北京邮电大学 | A kind of method, system and the equipment of intelligence water gauge reading |
| CN109764930A (en) * | 2018-12-27 | 2019-05-17 | 河海大学 | A kind of water gauge water level line visible detection method suitable for complex illumination condition |
| CN110956172A (en) * | 2019-11-18 | 2020-04-03 | 四创科技有限公司 | Water gauge identification method based on image semantic segmentation |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113221898A (en) * | 2021-04-16 | 2021-08-06 | 北京科技大学 | Automatic water level gauge reading method |
| CN113221898B (en) * | 2021-04-16 | 2022-02-15 | 北京科技大学 | Automatic water level gauge reading method |
| CN113141469A (en) * | 2021-04-21 | 2021-07-20 | 东莞市水利勘测设计院有限公司 | Anti-reflection method, system and device for night observation water gauge and readable storage medium |
| CN113657275A (en) * | 2021-08-16 | 2021-11-16 | 中国科学院空天信息创新研究院 | Automatic detection method for forest and grass fire points |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111476785B (en) | 2023-04-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111476785B (en) | Night infrared light-reflecting water gauge detection method based on position recording | |
| CN109902633B (en) | Abnormal event detection method and device based on fixed-position camera monitoring video | |
| CN115439424B (en) | Intelligent detection method for aerial video images of unmanned aerial vehicle | |
| CN111476827B (en) | Target tracking method, system, electronic device and storage medium | |
| CN111967498A (en) | Night target detection and tracking method based on millimeter wave radar and vision fusion | |
| CN113324864B (en) | Pantograph carbon slide plate abrasion detection method based on deep learning target detection | |
| CN112818988A (en) | Automatic reading identification method and system for pointer instrument | |
| CN111899288A (en) | Tunnel leakage water area detection and identification method based on infrared and visible light image fusion | |
| CN111965636A (en) | Night target detection method based on millimeter wave radar and vision fusion | |
| CN113688817A (en) | Instrument identification method and system for automatic inspection | |
| CN112560837A (en) | Reading method, device, equipment and storage medium of pointer instrument | |
| CN110610485A (en) | Ultra-high voltage transmission line channel hidden danger early warning method based on SSIM algorithm | |
| CN112927233A (en) | Marine laser radar and video combined target capturing method | |
| CN110363769B (en) | Image segmentation method for cantilever system of high-speed rail contact net supporting device | |
| CN108491857A (en) | A kind of multiple-camera target matching method of ken overlapping | |
| CN115019294A (en) | Pointer instrument reading identification method and system | |
| CN112017243B (en) | Medium visibility recognition method | |
| CN111597992B (en) | Scene object abnormity identification method based on video monitoring | |
| CN113763484A (en) | Ship target positioning and speed estimation method based on video image analysis technology | |
| CN112924037A (en) | Infrared body temperature detection system and detection method based on image registration | |
| CN109658405B (en) | Image data quality control method and system in crop live-action observation | |
| CN116310263A (en) | Pointer type aviation horizon instrument indication automatic reading implementation method | |
| CN112036391A (en) | Electronic inspection method and system based on shale gas field production station | |
| CN116958983A (en) | Pointer type pressure gauge automatic calibrator indication reading method based on machine vision | |
| CN111539329B (en) | Self-adaptive substation pointer instrument identification method |
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 |