CN119781083A - Group fog forecasting method based on improved double-brightness difference method and gray forecasting model - Google Patents
Group fog forecasting method based on improved double-brightness difference method and gray forecasting model Download PDFInfo
- Publication number
- CN119781083A CN119781083A CN202411567006.7A CN202411567006A CN119781083A CN 119781083 A CN119781083 A CN 119781083A CN 202411567006 A CN202411567006 A CN 202411567006A CN 119781083 A CN119781083 A CN 119781083A
- Authority
- CN
- China
- Prior art keywords
- gray
- forecasting
- visibility
- brightness difference
- prediction model
- 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
Classifications
-
- 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
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Image Processing (AREA)
Abstract
本发明涉及一种基于改进的双亮度差法和灰色预测模型的团雾预报方法，属于团雾的预测领域，包括以下步骤：(1)建立视频图像数据集；(2)图像灰度化；(3)图像滤波降噪；(4)特征区域分割提取；(5)计算能见度；(6)重复步骤(2)到步骤(5)，得到能见度的时间序列；(7)输入能见度时间序列，基于灰色预测模型实现对团雾的预报。本发明的优点是：使用改进的双亮度差法结合灰色预测模型，解决了现有的团雾预报方法中存在的气象设备分布稀疏、设备成本高和维护复杂等现状导致的监测范围受限，实现了对道路交通中团雾及时且精准的预报。
The present invention relates to a method for forecasting fog clusters based on an improved double brightness difference method and a gray prediction model, which belongs to the field of fog cluster prediction and includes the following steps: (1) establishing a video image data set; (2) graying the image; (3) filtering and denoising the image; (4) segmenting and extracting feature regions; (5) calculating visibility; (6) repeating steps (2) to (5) to obtain a time series of visibility; (7) inputting the visibility time series to forecast fog clusters based on a gray prediction model. The present invention has the advantages of using an improved double brightness difference method in combination with a gray prediction model to solve the problem of limited monitoring range caused by sparse distribution of meteorological equipment, high equipment cost and complex maintenance in existing fog cluster prediction methods, and to achieve timely and accurate forecasting of fog clusters in road traffic.
Description
Technical Field
The invention relates to a cluster fog forecasting method based on an improved double-brightness difference method and a gray forecasting model, and belongs to the field of cluster fog forecasting.
Background
The cluster fog is a weather phenomenon which has strong burst and small influence range but is extremely harmful to traffic safety, and visibility is generally used as an important parameter for cluster fog forecasting in the weather. The traditional cluster fog forecasting method mainly uses a visibility observer, a laser radar and other meteorological instruments for observation, but has the limitations of sparse equipment distribution, high equipment cost, complex maintenance and the like, so that the monitoring range is limited, local and transient characteristics of the cluster fog are difficult to capture, and the real-time and accurate requirements of cluster fog forecasting cannot be met. In recent years, the application of video monitoring and image processing technology enables capturing of the real-time dynamic of the generation of the mist, but different algorithms influence the accuracy of the visibility detection result, and monitoring and forecasting of the mist development trend cannot be achieved.
Disclosure of Invention
In order to solve the problems in the prior art, the invention provides a cluster fog forecasting method based on an improved double-brightness difference method and a gray forecasting model, which utilizes an improved image processing technology to analyze collected road monitoring video data so as to obtain a visibility time sequence, inputs the visibility time sequence into a GM (1, 1) gray forecasting model of an oscillation sequence, predicts the change trend of visibility along with time, realizes the forecasting of cluster fog and improves the safety of road traffic.
The invention provides a cluster fog forecasting method based on an improved double-brightness difference method and a gray forecasting model, which has the following technical scheme:
a cluster fog forecasting method based on an improved double-brightness difference method and a gray forecasting model comprises the following steps of (1) establishing a video image data set, (2) image graying, (3) image filtering and noise reduction, (4) feature area segmentation and extraction, (5) calculating visibility, (6) repeating the steps (2) to (5) to obtain a time sequence of the visibility, and (7) inputting the time sequence of the visibility to realize forecasting of the cluster fog based on the gray forecasting model.
The step (1) is specifically to decompose frame by frame based on video data collected by the road monitoring equipment to form a video image data set.
The step (2) is specifically as follows:
2.1 reading a color image in the video image dataset obtained in step (1);
2.2 the three red R, green G and blue B channels of each pixel in the color image are processed by a weighted average formula of GRAY=0.30r+0.59g+0.11b to become a GRAY GRAY single channel
2.3. And generating a gray image to realize image graying.
The step (3) specifically comprises the following steps:
and (3) filtering and denoising the gray image obtained in the step (2) by using a median filter.
In the step (4), the feature region is segmented by using a region growing method, and specifically includes the following steps:
4.1. selecting a pixel point as a seed in the sky area in the image obtained in the step (3);
4.2 merging similar pixel points in the neighborhood into the seed region through a threshold value, and performing neighborhood expansion;
4.3 repeating the step 4.1 and the step 4.2, combining the new pixel with the neighborhood thereof as a seed until all the pixels are contained in the characteristic region, and realizing region segmentation;
and 4.4, averaging the gray scale of each pixel point of the segmented sky area to obtain the gray scale Gt of the target object.
The step (5) specifically comprises the following steps:
Selecting two targets, wherein the gray scale of the two targets can be obtained in the step (4) and is respectively marked as Gt1 and Gt2, the distances from the targets to the camera lens are respectively L 1 and L 2,L2-L1, the distances are obtained through pavement marking, and the distances are substituted into a formula: Wherein, V d is visibility, G g1 and G g2 are background sky brightness of two targets respectively, and in order to reduce test error, the target of the selected area is consistent with the line of sight direction of the observation point, G g1＝Gg2 is caused, thus the visibility V d is calculated.
In the step (6), the method specifically comprises the following steps:
Repeating the steps (2) to (5), calculating the visibility of all images in the video image dataset, and simultaneously extracting the time information of each image to obtain a time sequence of the visibility.
The step (7) specifically comprises the steps of inputting the time sequence of the visibility obtained in the step (6) into a GM (1, 1) gray prediction model based on an oscillation sequence, respectively constructing a generation column, a matrix B and a data vector Y, and obtaining a time response sequence of the GM (1, 1) gray prediction model of the oscillation sequence, wherein the time response sequence is as follows: therefore, the visibility is predicted and analyzed, and the prediction of the mist is realized.
The method has the advantages that the problems that the existing cluster fog forecasting method is limited in monitoring range, local and transient characteristics of cluster fog are difficult to capture and the accuracy of visibility detection results is not ideal, the cluster fog development trend cannot be monitored and forecasted by combining an improved double-brightness difference method with a gray forecasting model, and the cluster fog in road traffic is timely and accurately forecasted are solved.
Drawings
FIG. 1 is a flow chart of the method for forecasting the mist of the present invention.
FIG. 2 is a schematic diagram showing the selection of seeds in a characteristic region by using a region growing method according to the cluster fog forecasting method of the present invention.
FIG. 3 is a graph showing the effect of region segmentation in the method for forecasting the mass fog.
Fig. 4 is a graph of distance measurement by pavement marking according to the method for forecasting the mass fog of the present invention.
FIG. 5 is a graph showing the result of the grey prediction model prediction in the method for forecasting the mass fog according to the present invention.
Detailed Description
The invention will be further described with reference to specific embodiments, and advantages and features of the invention will become apparent from the description. These examples are merely exemplary and do not limit the scope of the invention in any way. It will be understood by those skilled in the art that various changes and substitutions of details and forms of the technical solution of the present invention may be made without departing from the spirit and scope of the present invention, but these changes and substitutions fall within the scope of the present invention.
Referring to fig. 1 to 5, the invention relates to a cluster fog forecasting method based on an improved double-brightness difference method and a gray forecasting model, which comprises the following steps of (1) establishing a video image data set, (2) image graying, (3) image filtering and noise reduction, (4) feature region segmentation and extraction, (5) calculating visibility, (6) repeating the steps (2) to (5) to obtain a time sequence of the visibility, and (7) inputting the time sequence of the visibility and realizing the forecasting of the cluster fog based on the gray forecasting model.
Based on the setting of the steps, the following steps are realized:
The accuracy is improved, and the formation and development of the mist can be more accurately captured and predicted by combining a double brightness difference method and a gray prediction model.
And the real-time data processing is that the real-time processing and the establishment of the video image data set can timely respond to the environmental change and provide timely mass fog forecast.
The automation degree is high, manual intervention is reduced in an automatic image processing and data analysis process, and the forecasting efficiency and objectivity are improved.
The data utilization is full, namely, the information in the video image data is fully mined and utilized through the steps of graying, filtering noise reduction, feature region segmentation and the like.
The method is suitable for different environments and climatic conditions, and has strong adaptability and universal applicability.
The prediction timeliness is good, and through time sequence analysis, the prediction of the development trend of the mist can be provided, so that preventive measures can be taken in advance.
The step (1) is specifically to decompose frame by frame based on video data collected by the road monitoring equipment to form a video image data set.
The step (2) is specifically as follows:
2.1 reading a color image in the video image dataset obtained in step (1);
2.2, the three channels of red R, green G and blue B of each pixel in the color image are changed into a GRAY GRAY single channel after being processed by a weighted average formula, wherein GRAY=0.30R+0.59G+0.11B;
and 2.3, generating a gray image to realize image graying.
The step (3) is specifically that a median filter is used for filtering and noise reduction treatment of the gray level image obtained in the step (2).
In the step (4), the feature region is segmented by using a region growing method, and specifically includes the following steps:
4.1 as shown in fig. 2, selecting a pixel point as a seed (red square) in the sky area in the image obtained in the step (3);
4.2 merging similar pixel points in the neighborhood into a seed region through a threshold value to carry out neighborhood expansion, 4.3 repeating the steps 4.1 and 4.2, merging new pixels with the neighborhood by taking the new pixels as seeds until all the pixels are contained in a characteristic region, realizing region segmentation, and 4.4 averaging the gray scales of the pixel points in the segmented sky region to obtain the gray scale Gt of the target object, wherein the gray scales of the pixel points in the sky region are shown in figure 3.
The step (5) specifically comprises the following steps:
As shown in FIG. 4, two targets are selected, the gray scale of which can be obtained in the step (4) and is respectively marked as Gt1 and Gt2, the distances from the targets to the camera lens are respectively L1 and L2, and L2-L1 are obtained through road surface identification distances and are substituted into the formula: The Vd is visibility, gg1 and Gg 2 are background sky brightness of two targets respectively, and in order to reduce test errors, the directions of the vision of the targets in the selected area are consistent with the directions of the vision points, G g1＝Gg2 is enabled, so that the visibility Vd is calculated.
In the step (6), the method specifically comprises the following steps:
Repeating the steps (2) to (5), calculating the visibility of all images in the video image dataset, and simultaneously extracting the time information of each image to obtain a time sequence of the visibility.
The step (7) specifically comprises the steps of inputting the time sequence of the visibility obtained in the step (6) into a GM (1, 1) gray prediction model based on an oscillation sequence, respectively constructing a generation column, a matrix B and a data vector Y, and obtaining a time response sequence of the GM (1, 1) gray prediction model of the oscillation sequence, wherein the time response sequence is as follows: thus, the visibility is predicted and analyzed, the prediction of the mist is realized, and the prediction result is shown in figure 5.
The foregoing is only a preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art, who is within the scope of the present invention, should make equivalent substitutions or modifications according to the technical scheme of the present invention and the inventive concept thereof, and should be covered by the scope of the present invention.
Claims (8)
1. A cluster fog forecasting method based on an improved double-brightness difference method and a gray forecasting model is characterized by comprising the following steps of (1) establishing a video image data set, (2) image graying, (3) image filtering and noise reduction, (4) feature area segmentation and extraction, (5) calculating visibility, (6) repeating the steps (2) to (5) to obtain a time sequence of the visibility, and (7) inputting the time sequence of the visibility to realize forecasting of the cluster fog based on the gray forecasting model.
2. The method for forecasting the fog cluster based on the improved double brightness difference method and the gray prediction model according to claim 1, wherein the step (1) is specifically to decompose frame by frame based on video data collected by road monitoring equipment to form a video image data set.
3. The method for forecasting the cluster fog based on the improved dual brightness difference method and the gray prediction model according to claim 1, wherein the step (2) is specifically:
2.1 reading a color image in the video image dataset obtained in step (1);
2.2 the three red R, green G and blue B channels of each pixel in the color image are processed by a weighted average formula of GRAY=0.30R+0.59G+0.11B to become a GRAY GRAY single channel
2.3. And generating a gray image to realize image graying.
4. The method for forecasting the cluster fog based on the improved double brightness difference method and the gray prediction model according to claim 1, wherein the step (3) is specifically:
and (3) filtering and denoising the gray image obtained in the step (2) by using a median filter.
5. The method for forecasting the cluster fog based on the improved dual brightness difference method and the gray prediction model according to claim 1, wherein in the step (4), the feature area is segmented by using an area growing method, and the method specifically comprises the following steps:
4.1. selecting a pixel point as a seed in the sky area in the image obtained in the step (3);
4.2 merging similar pixel points in the neighborhood into the seed region through a threshold value, and performing neighborhood expansion;
4.3 repeating the step 4.1 and the step 4.2, combining the new pixel with the neighborhood thereof as a seed until all the pixels are contained in the characteristic region, and realizing region segmentation;
and 4.4, averaging the gray scale of each pixel point of the segmented sky area to obtain the gray scale Gt of the target object.
6. The method for forecasting the cluster fog based on the improved double brightness difference method and the gray prediction model according to claim 1, wherein in the step (5), the method specifically comprises the following steps:
Selecting two targets, wherein the gray scale of the two targets can be obtained in the step (4) and is respectively marked as Gt1 and Gt2, the distances from the targets to the camera lens are respectively L 1 and L 2,L2-L1, the distances are obtained through pavement marking, and the distances are substituted into a formula: Wherein, V d is visibility, G g1 and G g2 are background sky brightness of two targets respectively, and in order to reduce test error, the target of the selected area is consistent with the line of sight direction of the observation point, G g1＝Gg2 is caused, thus the visibility V d is calculated.
7. The method for forecasting the cluster fog based on the improved double brightness difference method and the gray prediction model according to claim 1, wherein in the step (6), the method specifically comprises the following steps:
Repeating the steps (2) to (5), calculating the visibility of all images in the video image dataset, and simultaneously extracting the time information of each image to obtain a time sequence of the visibility.
8. The method for forecasting the cluster fog based on the improved double brightness difference method and the gray prediction model according to claim 1, wherein in the step (7), the method specifically comprises the following steps of inputting the time sequence of the visibility obtained in the step (6) into the GM (1, 1) gray prediction model based on the oscillation sequence, respectively constructing a generating column, a matrix B and a data vector Y, and obtaining a time response sequence of the GM (1, 1) gray prediction model of the oscillation sequence as follows: therefore, the visibility is predicted and analyzed, and the prediction of the mist is realized.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411567006.7A CN119781083B (en) | 2024-11-05 | 2024-11-05 | A method for cluster fog prediction based on improved double brightness difference method and grey prediction model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411567006.7A CN119781083B (en) | 2024-11-05 | 2024-11-05 | A method for cluster fog prediction based on improved double brightness difference method and grey prediction model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN119781083A true CN119781083A (en) | 2025-04-08 |
| CN119781083B CN119781083B (en) | 2025-09-30 |
Family
ID=95239950
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202411567006.7A Active CN119781083B (en) | 2024-11-05 | 2024-11-05 | A method for cluster fog prediction based on improved double brightness difference method and grey prediction model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN119781083B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106871084A (en) * | 2017-03-24 | 2017-06-20 | 广州市浩洋电子股份有限公司 | A kind of stage lighting with movable light source system |
| CN109377726A (en) * | 2018-10-29 | 2019-02-22 | 江苏大学 | An accurate warning and induction system and method for expressway fog mass based on the Internet of Vehicles |
| CN109934780A (en) * | 2019-02-21 | 2019-06-25 | 北京以萨技术股份有限公司 | A kind of traffic surveillance videos defogging method based on dark primary priori |
| CN214369912U (en) * | 2021-04-01 | 2021-10-08 | 东莞市力拓五金压铸有限公司 | Aluminum accessory for lamp decoration |
| CN113484255A (en) * | 2021-08-03 | 2021-10-08 | 南京信息工程大学滨江学院 | Modified visibility estimation method based on near-infrared double-spectrum real-time imaging and distance |
| CN116188316A (en) * | 2023-03-06 | 2023-05-30 | 西安电子科技大学 | A water area defogging method based on fog density perception |
-
2024
- 2024-11-05 CN CN202411567006.7A patent/CN119781083B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106871084A (en) * | 2017-03-24 | 2017-06-20 | 广州市浩洋电子股份有限公司 | A kind of stage lighting with movable light source system |
| CN109377726A (en) * | 2018-10-29 | 2019-02-22 | 江苏大学 | An accurate warning and induction system and method for expressway fog mass based on the Internet of Vehicles |
| CN109934780A (en) * | 2019-02-21 | 2019-06-25 | 北京以萨技术股份有限公司 | A kind of traffic surveillance videos defogging method based on dark primary priori |
| CN214369912U (en) * | 2021-04-01 | 2021-10-08 | 东莞市力拓五金压铸有限公司 | Aluminum accessory for lamp decoration |
| CN113484255A (en) * | 2021-08-03 | 2021-10-08 | 南京信息工程大学滨江学院 | Modified visibility estimation method based on near-infrared double-spectrum real-time imaging and distance |
| CN116188316A (en) * | 2023-03-06 | 2023-05-30 | 西安电子科技大学 | A water area defogging method based on fog density perception |
Non-Patent Citations (3)
| Title |
|---|
| "时间序列分析----灰度模型", HTTPS://BLOG.CSDN.NET/U014106644/ARTICLE/DETAILS/90405842, 31 May 2019 (2019-05-31) * |
| 吕伟涛: "双亮度差方法测量白天气象能见度时的误差分析", 应用气象学报, vol. 16, no. 5, 31 October 2005 (2005-10-31), pages 619 - 628 * |
| 曹蓉: "大雾演化规律探究及变化趋势预测", 数学的实践与认识, vol. 51, no. 23, 31 December 2021 (2021-12-31), pages 275 - 283 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN119781083B (en) | 2025-09-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112149543B (en) | A computer vision-based construction dust identification system and method | |
| CN103630496B (en) | Based on the traffic video visibility detecting method of road surface apparent brightness and least square method | |
| CN119246518B (en) | Bridge surface defect detection method and system based on laser point cloud and visual image fusion | |
| CN101334844A (en) | Key Feature Extraction Method for Flotation Froth Image Analysis | |
| CN115937138B (en) | A bridge crack identification and detection system and method based on drone | |
| CN111126183A (en) | Method for detecting damage of building after earthquake based on near-ground image data | |
| CN115639248A (en) | System and method for detecting quality of building outer wall | |
| CN116703787B (en) | Building construction safety risk early warning method and system | |
| CN113033386A (en) | High-resolution remote sensing image-based transmission line channel hidden danger identification method and system | |
| CN113781436B (en) | High-voltage switch state monitoring method based on camera and angle measurement | |
| CN109684986B (en) | Vehicle analysis method and system based on vehicle detection and tracking | |
| CN102393902A (en) | Vehicle color detection method based on H_S two-dimensional histogram and regional color matching | |
| CN112884795A (en) | Power transmission line inspection foreground and background segmentation method based on multi-feature significance fusion | |
| CN106971396A (en) | Ice sheet freeze thawing detection method based on super-pixel | |
| CN112927233A (en) | Marine laser radar and video combined target capturing method | |
| CN119470967A (en) | A method for regional monitoring of water surface velocity based on intelligent visual analysis | |
| CN111104850A (en) | A method and system for automatic extraction of buildings from remote sensing images based on residual network | |
| CN113628251B (en) | Smart hotel terminal monitoring method | |
| CN116758081B (en) | Unmanned aerial vehicle road and bridge inspection image processing method | |
| CN107705330B (en) | Intelligent estimation method of visibility based on road camera | |
| CN119781083B (en) | A method for cluster fog prediction based on improved double brightness difference method and grey prediction model | |
| CN114140698B (en) | A water system information extraction algorithm based on FasterR-CNN | |
| CN111652935B (en) | Positioning system and method for railway train bracket | |
| CN114463678A (en) | Rainfall type identification method using camera video image | |
| CN116563774B (en) | Automatic identification method and system for illegal mining of cultivated land in China |
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 |