CN114332777B - High-altitude parabolic detection method and device - Google Patents
High-altitude parabolic detection method and device Download PDFInfo
- Publication number
- CN114332777B CN114332777B CN202210217690.0A CN202210217690A CN114332777B CN 114332777 B CN114332777 B CN 114332777B CN 202210217690 A CN202210217690 A CN 202210217690A CN 114332777 B CN114332777 B CN 114332777B
- Authority
- CN
- China
- Prior art keywords
- image
- frame
- current frame
- parabolic
- integral
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Landscapes
- Image Analysis (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
The invention discloses a high-altitude parabolic detection method, which comprises the following steps: acquiring difference images of image channels of a current frame and a previous frame of a monitoring video, and combining the difference images of the image channels into a first frame difference image of the current frame; integrating the first frame difference image of the current frame and the first frame difference images of a plurality of previous frames according to a first period to obtain a first integral image of the current frame; taking a difference image of the first integral image of the current frame and the first integral image of the previous frame as a second frame difference image of the current frame, and performing weighted superposition on the second frame difference image of the current frame and the second frame difference images of a plurality of previous frames according to a second period to obtain a second integral image of the current frame; and judging whether high-altitude parabolic motion occurs according to the second integral image of the current frame. The invention also discloses a high-altitude parabolic detection device. Compared with the prior art, the high-altitude parabolic detection method has the advantages of higher detection precision and better detection real-time property.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a high-altitude parabolic detection method.
Background
With the gradual deepening of the urbanization process, more and more high-rise buildings appear in cities, and the high-altitude parabolic behavior is also more and more concerned by the society. As the incident places are high-altitude floors, witnesses are few, and the parabolic time is short, law enforcement departments are difficult to follow the legal responsibility of the parabolic persons. Therefore, the automatic detection of the high altitude parabola by adopting the monitoring video becomes the most feasible means.
The high-altitude parabolic detection based on the surveillance video essentially belongs to a multi-target tracking task, so that the high-altitude parabolic detection can be realized by adopting the conventional multi-target tracking algorithm. However, compared with other task scenes, the high-altitude parabolic detection has the characteristics of extremely small target, high falling speed, fewer track points shot by a common monitoring camera and the like. The characteristics limit the practical effect of the existing multi-target tracking algorithm in high-altitude parabolic detection.
In the prior art, a detection method, a foreground modeling method or a frame difference foreground detection method is generally used for detecting the position of a parabola in an image, then a tracker is used for tracking a target, and whether a track accords with characteristics of the parabola is judged, for example, a target tracking algorithm combining maximum foreground detection and Kalman filtering is used. The performance requirements of the detection method are severe, and meanwhile, semantic information does not exist for some flying insects and flying birds, namely, only some detection frames are tracked, whether the detected birds or the insects are detected or false detection caused by rain and snow weather is not known, so that the high-altitude parabolic scheme based on the detection tracking mode is extremely difficult to control for abnormal judgment.
In order to overcome the defects of the above schemes, some researchers use a moving object detection scheme based on inter-frame difference to perform high-altitude parabolic detection, that is, image processing and analysis are performed on a frame difference image of a monitoring video to obtain a high-altitude parabolic detection result, for example, the technical schemes disclosed in chinese patent applications CN111768431A and CN 113139478A. Although the scheme can improve the real-time performance of detection to a certain extent, the problems of branch shaking, light flickering and the like cannot be eliminated, and a considerable false detection rate still exists.
Disclosure of Invention
The invention aims to solve the technical problem of overcoming the defects of the prior art and provides a simple, direct, visual and clear high-altitude parabolic detection method which is higher in detection precision and better in detection real-time performance.
The invention specifically adopts the following technical scheme to solve the technical problems:
a high altitude parabolic detection method comprises the following steps:
acquiring difference images of image channels of a current frame and a previous frame of a monitoring video, and combining the difference images of the image channels into a first frame difference image of the current frame;
integrating the first frame difference image of the current frame and the first frame difference images of a plurality of previous frames according to a first period to obtain a first integral image of the current frame;
taking a difference image of the first integral image of the current frame and the first integral image of the previous frame as a second frame difference image of the current frame, and performing weighted superposition on the second frame difference image of the current frame and the second frame difference images of a plurality of previous frames according to a second period to obtain a second integral image of the current frame;
and judging whether high-altitude parabolic motion occurs according to the second integral image of the current frame.
Preferably, the determining whether a high altitude parabola occurs according to the second integral image of the current frame specifically includes: and inputting the second integral image of the current frame into a pre-trained classifier, wherein the output of the classifier is the judgment result of whether high altitude parabolic occurs.
Further preferably, the classifier is a repvgg network with attention mechanism.
Preferably, the method of merging comprises: taking the maximum pixel value in each image channel difference image as the pixel value of the first frame difference image; or, taking the pixel value mean value in each image channel difference image as the pixel value of the first frame difference image.
Still further, the method of merging further comprises: and setting the pixel value smaller than the preset threshold value in the first frame difference image as 0.
Preferably, in the weighted overlap-add, the first and second weights are, in the weighted overlap-add,tweight of the second frame difference map of the current frame at the moment
Determined according to the following formula:
in the formula (I), the compound is shown in the specification,T2 is a second period of time, 2,eis a natural constant and is a natural constant,γa control parameter greater than 0.
Further, the method further comprises:
and extracting a parabolic track and/or a parabolic position by an image processing method for the second integral image which is judged to generate the high-altitude parabola.
Based on the same inventive concept, the following technical scheme can be obtained:
a high altitude parabolic detection apparatus comprising:
the first frame difference image acquisition module is used for acquiring difference images of each image channel of a current frame and a previous frame of the monitoring video and combining the difference images of each image channel into a first frame difference image of the current frame;
the first integration module is used for integrating the first frame difference image of the current frame and the first frame difference images of a plurality of previous frames according to a first period to obtain a first integration image of the current frame;
the second integral module is used for taking a difference image of the first integral image of the current frame and the first integral image of the previous frame as a second frame difference image of the current frame, and performing weighted superposition on the second frame difference image of the current frame and the second frame difference images of a plurality of previous frames according to a second period to obtain a second integral image of the current frame;
and the parabolic judging module is used for judging whether high-altitude parabolic occurs according to the second integral image of the current frame.
Preferably, the parabola judgment module is a pre-trained classifier, the input of the classifier is the second integral image of the current frame, and the output of the classifier is the judgment result of whether high altitude parabola occurs.
Further preferably, the classifier is a repvgg network with attention mechanism.
Preferably, the method of merging comprises: taking the maximum pixel value in each image channel difference image as the pixel value of the first frame difference image; or, taking the pixel value mean value in each image channel difference image as the pixel value of the first frame difference image.
Still further, the method of merging further comprises: and setting the pixel value smaller than the preset threshold value in the first frame difference image as 0.
Preferably, in the weighted overlap-add, the first and second weights are, in the weighted overlap-add,tweight of the second frame difference map of the current frame at the moment
Determined according to the following formula:
in the formula (I), the compound is shown in the specification,T2 is the firstThe period of the cycle is set to be,eis a natural constant and is a natural constant,γa control parameter greater than 0.
Further, the apparatus further comprises:
and the information extraction module is used for extracting the parabolic track and/or the parabolic position for the second integral image which is judged to generate the high-altitude parabola through an image processing method.
Compared with the prior art, the technical scheme of the invention has the following beneficial effects:
the invention uses an end-to-end processing mode, the high-altitude parabolic detection method is clear and simple, and the detection result is more visual;
according to the method, the complete path of the parabola can be directly seen based on the second integral image obtained by frame difference image processing, repeated reference of the parabola video is not needed, and the parabola generation can be judged conveniently in a machine classification mode or information such as the parabola track and the parabola position can be obtained conveniently in a simple image processing mode; the method and the device can effectively eliminate false detection caused by the problems of branch shaking, light flickering and the like, and have better detection accuracy compared with the conventional parabolic detection scheme based on the frame difference image.
Drawings
FIG. 1 shows a first integral image S obtained in the examplet；
FIG. 2 is a second integral image M obtained in the examplet；
Fig. 3 is a comparison diagram of the detection process, in which the second integral image, the CAM attention map and the current frame image are sequentially arranged from left to right.
Detailed Description
Aiming at the defects of the prior art, the solution idea of the invention is to obtain a second integral image by carrying out twice difference and twice integral processing on the image frame of the monitoring video and carry out high-altitude parabolic detection based on the second integral image; the complete path of the parabola can be directly seen through the second integral image, the parabola video does not need to be repeatedly consulted, and the generation of the parabola can be conveniently judged in a machine classification mode or information such as the parabola track and the parabola position can be conveniently obtained in a simple image processing mode; the method and the device can effectively eliminate false detection caused by the problems of branch shaking, light flickering and the like, and have better detection accuracy compared with the conventional parabolic detection scheme based on the frame difference image.
The invention provides a high-altitude parabolic detection method, which specifically comprises the following steps:
acquiring difference images of image channels of a current frame and a previous frame of a monitoring video, and combining the difference images of the image channels into a first frame difference image of the current frame;
integrating the first frame difference image of the current frame and the first frame difference images of a plurality of previous frames according to a first period to obtain a first integral image of the current frame;
taking a difference image of the first integral image of the current frame and the first integral image of the previous frame as a second frame difference image of the current frame, and performing weighted superposition on the second frame difference image of the current frame and the second frame difference images of a plurality of previous frames according to a second period to obtain a second integral image of the current frame;
and judging whether high-altitude parabolic motion occurs according to the second integral image of the current frame.
The invention provides a high-altitude parabolic detection device, which comprises:
the first frame difference image acquisition module is used for acquiring difference images of each image channel of a current frame and a previous frame of the monitoring video and combining the difference images of each image channel into a first frame difference image of the current frame;
the first integration module is used for integrating the first frame difference image of the current frame and the first frame difference images of a plurality of previous frames according to a first period to obtain a first integration image of the current frame;
the second integral module is used for taking a difference image of the first integral image of the current frame and the first integral image of the previous frame as a second frame difference image of the current frame, and performing weighted superposition on the second frame difference image of the current frame and the second frame difference images of a plurality of previous frames according to a second period to obtain a second integral image of the current frame;
and the parabolic judging module is used for judging whether high-altitude parabolic occurs according to the second integral image of the current frame.
Preferably, the determining whether a high altitude parabola occurs according to the second integral image of the current frame specifically includes: and inputting the second integral image of the current frame into a pre-trained classifier, wherein the output of the classifier is the judgment result of whether high altitude parabolic occurs.
Further preferably, the classifier is a repvgg network with attention mechanism.
Preferably, the method of merging comprises: taking the maximum pixel value in each image channel difference image as the pixel value of the first frame difference image; or, taking the pixel value mean value in each image channel difference image as the pixel value of the first frame difference image.
Still further, the method of merging further comprises: and setting the pixel value smaller than the preset threshold value in the first frame difference image as 0.
Preferably, in the weighted overlap-add, the first and second weights are, in the weighted overlap-add,tweight of the second frame difference map of the current frame at the moment
Determined according to the following formula:
in the formula (I), the compound is shown in the specification,T2 is a second period of time, 2,eis a natural constant and is a natural constant,γa control parameter greater than 0.
Further, the method further comprises:
and extracting a parabolic track and/or a parabolic position by an image processing method for the second integral image which is judged to generate the high-altitude parabola.
For the public understanding, the technical scheme of the invention is explained in detail by a specific embodiment and the accompanying drawings:
the high-altitude parabolic detection process in the embodiment is specifically as follows:
1. the camera acquires a video stream, which may be represented as a combination of a series of pictures (I)o ... It) In which ItIs shown astThe picture of the current frame at that moment. The condition that the camera is good in light is commonIn the case of an RGB video stream, an IR video stream is output at night when the light is poor, where the image can be represented as a rectangular block of data with dimensions (w, h, c),
is shown astThe number of channels for a full-color image is typically 3 for the c-th image channel of the current frame at time instant.
2. Separate all image channels, pair
The following operations are carried out
Obtaining a difference image of the image frames at the t moment and the t-1 moment, namely obtaining a first frame difference image of the current frame at the t momentdiff t WhereinabsThe function is a return absolute value function, and the max function is a maximum function. Respectively calculating difference images of each image channel and then merging, wherein the merging mode can be that the maximum pixel value in the difference images of each image channel is used as the pixel value of the first frame of difference image; or, taking the pixel value mean value in each image channel difference image as the pixel value of the first frame difference image; in this embodiment, the maximum value on each image channel is taken as the final response. In order to further eliminate the noise interference, a threshold value may be set in advance, and a pixel value smaller than the threshold value in the first frame difference map may be set to 0.
3. Due to the particularity of the high-altitude parabola, the high-altitude parabola generally falls from top to bottom in the picture, and the complete parabola falling track image can be obtained by performing fixed-period integration on the image variation. Specifically, an image integration duration T1 is set, and a first frame difference image of the current frame and first frame difference images of previous frames are integrated according to T1 to obtain a first integral image of the current frame at time T
(ii) a The first integral image S obtainedtAn example of the method is shown in fig. 1, and it can be found that the method can accurately and intuitively reflect the moving target track, but cannot eliminate the problems of treeing, light flickering and the like, and in fig. 1, treeing which shakes in the wind is retained. The specific value range of the period T1 is determined flexibly according to the camera deployment position, the camera focal length, the floor and other factors, the T1 value is slightly higher than the time from the time when the parabola enters the picture to the time when the parabola leaves the picture, and the more common setting range is 0.5s-10 s.
4. And differentiating the first integral image sequence to obtain a differential image of the first integral image, namely taking a difference image of the first integral image of the current frame and the first integral image of the previous frame as a second frame difference image D of the current framet(ii) a Then, according to a period T2, the second frame difference image of the current frame and the second frame difference images of a plurality of previous frames are weighted and superposed to obtain a second integral image M of the current framet：
By calculating the second integral image MtCan effectively eliminate the problems of branch shaking, light flickering and the like, thereby being capable of more accurately grabbing high-altitude parabolic features, wherein the weight value
Representing time decay coefficients for controlling parameters concerning recently occurring events
Greater than 0 for adjusting the degree of forgetting for the history. As shown in FIG. 2, the treeing of FIG. 1 is shakingThe branches have been completely removed in fig. 2. The period T2 may be the same as or different from the period T1, and may be set as needed.
5. Integrating the second integral image MtSending the object to a pre-trained classifier for classification, and judging whether the object is a high-altitude object; the classifier can be an existing feedforward neural network or a deep learning network such as CNN, VGG, ResNet, and the like, and the classifier in this embodiment adopts a repvgg network with an attention mechanism. The classification network sample data constructs a data set by using a simulation and collection mode, a falling mode of a parabolic track is simulated, a classification model is trained to be fitted according to the data, and effective judgment can be made on the conditions of severe weather and abnormal conditions (such as birds and winged insects). And drawing an ROC curve through the test set, and selecting proper recall rate and false detection rate according to scene needs to obtain a classification model threshold value.
6. As shown in fig. 3 (the second integral image, the CAM attention map and the current frame image are sequentially arranged from left to right), the pleasing region is drawn for the repvgg, the repvgg attention region is drawn by using the CAM (class Activation mapping), and the attention map can be used for marking the suspicious region of high altitude parabola generation.
7. Repeating the steps 1-6 to obtain a second integral image of the current frame, and judging whether each frame has a high-altitude parabola or not; setting the sensitivity window time to TwThe detection result within the window at time t can be expressed as
. The sensitivity is marked sens, and the output result of the repvgg classification model at the time t is parabolic proportion. And when the proportion of the classification result in the window time is greater than a set threshold, judging that high-altitude parabolic motion occurs, otherwise, judging that the system is normal.
8. And extracting a parabolic track and/or a parabolic position by an image processing method for the second integral image which is judged to generate the high-altitude parabola. Because the obtained second integral image eliminates the problems of treeing, light flicker and the like, and the parabolic track is clearly and completely reflected, a simple image processing method can be adopted to extract the complete parabolic track from the second integral image and obtain accurate parabolic position information (specifically, floors, windows and the like of thrown objects) according to the coordinate comparison between the parabolic track and the original video frame, for example, the following method can be adopted:
setting a threshold value tb for the second integral image obtained by the method to carry out binarization, setting the threshold value to be 255 when each pixel in the second integral image is larger than tb, and otherwise, setting the threshold value to be 0 to obtain a binary image Mb;
using a contour searching mode for the binary image Mb, for example, using a findContours function in opencv to perform contour searching, obtaining center points of all contours, removing abnormal contours through contour areas, obtaining a parabolic track contour center point set, and sorting points in the point set from high to low according to a y axis to obtain a point set C;
the point set C is sequenced from top to bottom, so the starting point is the starting point of the high altitude parabola, the number of the floor of the building can be determined by various methods, for example, the number of the floor is marked in advance, and the parabola position can be confirmed by inquiring the mark closest to the starting point; or fitting the coordinate relation between the floor and the image by means of perspective transformation and the like, and inputting the coordinates of the starting point to obtain the floor number.
Claims (14)
1. A high altitude parabolic detection method is characterized by comprising the following steps:
acquiring difference images of image channels of a current frame and a previous frame of a monitoring video, and combining the difference images of the image channels into a first frame difference image of the current frame;
integrating the first frame difference image of the current frame and the first frame difference images of a plurality of previous frames according to a first period to obtain a first integral image of the current frame;
taking a difference image of the first integral image of the current frame and the first integral image of the previous frame as a second frame difference image of the current frame, and performing weighted superposition on the second frame difference image of the current frame and the second frame difference images of a plurality of previous frames according to a second period to obtain a second integral image of the current frame;
and judging whether high-altitude parabolic motion occurs according to the second integral image of the current frame.
2. The high-altitude parabolic detection method according to claim 1, wherein the determining whether the high-altitude parabolic occurs according to the second integral image of the current frame specifically comprises: and inputting the second integral image of the current frame into a pre-trained classifier, wherein the output of the classifier is the judgment result of whether high altitude parabolic occurs.
3. The high altitude parabola detection method as claimed in claim 2, wherein the classifier is a repvgg network with attention mechanism.
4. The high altitude parabolic detection method according to claim 1, wherein the combining method comprises: taking the maximum pixel value in each image channel difference image as the pixel value of the first frame difference image; or, taking the pixel value mean value in each image channel difference image as the pixel value of the first frame difference image.
5. The high altitude parabolic detection method according to claim 4, wherein the combining method further comprises: and setting the pixel value smaller than the preset threshold value in the first frame difference image as 0.
6. The high altitude parabolic detection method according to claim 1, wherein in the weighted overlap-add,tweight of the second frame difference map of the current frame at the moment
Determined according to the following formula:
in the formula (I), the compound is shown in the specification,T2 is a second period of time, 2,eis a natural constant and is a natural constant,γa control parameter greater than 0.
7. The high altitude parabolic detection method according to claim 1, further comprising:
and extracting a parabolic track and/or a parabolic position by an image processing method for the second integral image which is judged to generate the high-altitude parabola.
8. A high altitude parabolic detection device, characterized by comprising:
the first frame difference image acquisition module is used for acquiring difference images of each image channel of a current frame and a previous frame of the monitoring video and combining the difference images of each image channel into a first frame difference image of the current frame;
the first integration module is used for integrating the first frame difference image of the current frame and the first frame difference images of a plurality of previous frames according to a first period to obtain a first integration image of the current frame;
the second integral module is used for taking a difference image of the first integral image of the current frame and the first integral image of the previous frame as a second frame difference image of the current frame, and performing weighted superposition on the second frame difference image of the current frame and the second frame difference images of a plurality of previous frames according to a second period to obtain a second integral image of the current frame;
and the parabolic judging module is used for judging whether high-altitude parabolic occurs according to the second integral image of the current frame.
9. The high altitude parabolic detection device according to claim 8, wherein the parabolic determination module is a pre-trained classifier, an input of the classifier is a second integral image of a current frame, and an output of the classifier is a determination result of whether high altitude parabolic occurs.
10. The high altitude parabola detection apparatus of claim 9, wherein the classifier is a repvgg network with attention mechanism.
11. The high altitude parabolic detection apparatus of claim 8, wherein the method of combining comprises: taking the maximum pixel value in each image channel difference image as the pixel value of the first frame difference image; or, taking the pixel value mean value in each image channel difference image as the pixel value of the first frame difference image.
12. The high altitude parabolic detection apparatus of claim 11, wherein the method of combining further comprises: and setting the pixel value smaller than the preset threshold value in the first frame difference image as 0.
13. The high altitude parabola detection apparatus of claim 8, wherein in said weighted overlap-add,tweight of the second frame difference map of the current frame at the moment
Determined according to the following formula:
in the formula (I), the compound is shown in the specification,T2 is a second period of time, 2,eis a natural constant and is a natural constant,γa control parameter greater than 0.
14. The high altitude parabola detection apparatus of claim 8, further comprising:
and the information extraction module is used for extracting the parabolic track and/or the parabolic position for the second integral image which is judged to generate the high-altitude parabola through an image processing method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210217690.0A CN114332777B (en) | 2022-03-08 | 2022-03-08 | High-altitude parabolic detection method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210217690.0A CN114332777B (en) | 2022-03-08 | 2022-03-08 | High-altitude parabolic detection method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114332777A CN114332777A (en) | 2022-04-12 |
| CN114332777B true CN114332777B (en) | 2022-05-17 |
Family
ID=81031222
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210217690.0A Active CN114332777B (en) | 2022-03-08 | 2022-03-08 | High-altitude parabolic detection method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114332777B (en) |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7627139B2 (en) * | 2002-07-27 | 2009-12-01 | Sony Computer Entertainment Inc. | Computer image and audio processing of intensity and input devices for interfacing with a computer program |
| CN112291478B (en) * | 2020-11-05 | 2022-08-30 | 北京软通智慧科技有限公司 | Method, device and equipment for monitoring high-altitude falling object and storage medium |
| CN113409362B (en) * | 2021-06-29 | 2023-02-21 | 深圳市商汤科技有限公司 | High altitude parabolic detection method and device, equipment and computer storage medium |
-
2022
- 2022-03-08 CN CN202210217690.0A patent/CN114332777B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN114332777A (en) | 2022-04-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10950005B1 (en) | Dynamic calibration of surveillance devices | |
| WO2021088300A1 (en) | Rgb-d multi-mode fusion personnel detection method based on asymmetric double-stream network | |
| Craye et al. | Spatio-temporal semantic segmentation for drone detection | |
| CN104463196B (en) | A kind of weather phenomenon recognition methods based on video | |
| CN110660222B (en) | Intelligent environment-friendly electronic snapshot system for black-smoke road vehicle | |
| CN110490043A (en) | A kind of forest rocket detection method based on region division and feature extraction | |
| KR102392822B1 (en) | Device of object detecting and tracking using day type camera and night type camera and method of detecting and tracking object | |
| CN110956146B (en) | Road background modeling method and device, electronic equipment and storage medium | |
| US11012683B1 (en) | Dynamic calibration of surveillance devices | |
| CN114639075B (en) | Method and system for identifying falling object of high altitude parabola and computer readable medium | |
| CN106339657A (en) | Straw incineration monitoring method and device based on monitoring video | |
| US11756303B2 (en) | Training of an object recognition neural network | |
| CN109948474A (en) | AI thermal imaging all-weather intelligent monitoring method | |
| CN114202646A (en) | Infrared image smoking detection method and system based on deep learning | |
| CN108830161B (en) | Smog identification method based on video stream data | |
| CN112613336A (en) | Method and device for generating an object classification of an object | |
| CN115035470A (en) | Low, small and slow target identification and positioning method and system based on mixed vision | |
| Lalonde et al. | A system to automatically track humans and vehicles with a PTZ camera | |
| CN116258940A (en) | Small target detection method for multi-scale features and self-adaptive weights | |
| Li et al. | Weak moving object detection in optical remote sensing video with motion-drive fusion network | |
| CN1556506A (en) | Intelligent alarming treatment method of video frequency monitoring system | |
| CN110211159A (en) | A kind of aircraft position detection system and method based on image/video processing technique | |
| CN114332777B (en) | High-altitude parabolic detection method and device | |
| CN111476314B (en) | Fuzzy video detection method integrating optical flow algorithm and deep learning | |
| KR102491684B1 (en) | Control method of a surveillance camera with a sunshield and control method thereof |
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 | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 210000 Longmian Avenue 568, High-tech Park, Jiangning District, Nanjing City, Jiangsu Province Patentee after: Xiaoshi Technology (Jiangsu) Co.,Ltd. Address before: 210000 Longmian Avenue 568, High-tech Park, Jiangning District, Nanjing City, Jiangsu Province Patentee before: NANJING ZHENSHI INTELLIGENT TECHNOLOGY Co.,Ltd. |
|
| CP01 | Change in the name or title of a patent holder |