CN113905178B - Environment automatic sensing cruising method based on high-altitude holder - Google Patents
Environment automatic sensing cruising method based on high-altitude holder Download PDFInfo
- Publication number
- CN113905178B CN113905178B CN202111185406.8A CN202111185406A CN113905178B CN 113905178 B CN113905178 B CN 113905178B CN 202111185406 A CN202111185406 A CN 202111185406A CN 113905178 B CN113905178 B CN 113905178B
- Authority
- CN
- China
- Prior art keywords
- value
- array
- image
- altitude
- cruising
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
- H04N23/67—Focus control based on electronic image sensor signals
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C1/00—Registering, indicating or recording the time of events or elapsed time, e.g. time-recorders for work people
- G07C1/20—Checking timed patrols, e.g. of watchman
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
- H04N23/64—Computer-aided capture of images, e.g. transfer from script file into camera, check of taken image quality, advice or proposal for image composition or decision on when to take image
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
- H04N23/695—Control of camera direction for changing a field of view, e.g. pan, tilt or based on tracking of objects
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/80—Camera processing pipelines; Components thereof
-
- 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
Abstract
The invention discloses an environment automatic perception cruising method based on a high-altitude tripod head, which comprises a first stage and a second stage; stage one includes: initializing a high-altitude pan-tilt camera, and setting PTZ values of the high-altitude pan-tilt camera to be minimum values; triple cyclic scanning is carried out on the PTZ, and a photo is taken, so that an arithmetic sequence image is formed; selecting the most suitable Z value image from the arithmetic sequence images, and storing the most suitable Z value image; the second stage comprises the following steps: under the condition that the Z value is unchanged, removing images with redundancy exceeding the repetition proportion according to the distance of P, T value; and picking out the image sequence with the nearest PZ value distance, and re-sequencing and outputting the PTZ value of the image sequence as an intelligent cruising route. The invention can automatically sense the distance of the main object in the image, thereby automatically realizing intelligent cruising according to the distance and the recently of the main object.
Description
Technical Field
The invention relates to the technical field of image processing information, in particular to an environment automatic perception cruising method based on a high-altitude tripod head.
Background
The high-altitude tripod head camera is characterized in that high-multiple video equipment and supporting facilities are arranged at a position 50-100 meters or more away from the ground (such as a forest grassland observation tower, a forest fire video monitoring tripod head and the like) so as to realize video image monitoring in a range of several kilometers or even more, and the high-altitude tripod head camera is a whistle of a forest in China. According to the national forest grass bureau forest fire 2016-2025 program, the national forest region video monitoring coverage rate will reach over 95% in 2025, with a significant portion of the monitoring coming from Gao Kongyun cameras. The traditional high-altitude tripod head camera can only mechanically cruise according to a manually set route of people, does not have an intelligent environment sensing function and cannot sense the distance of objects in the environment, so that a reasonable cruising route is automatically planned; therefore, the advantages of the high-power telescopic function are not exerted, and the target identification and tracking in the video monitoring in a large range cannot be really solved.
The traditional camera shooting systems all have an automatic focusing function, and more advanced systems also have the functions of single automatic focusing (ONE SHOT), artificial intelligent automatic focusing (AI FOCUS), artificial intelligent SERVO automatic focusing (AI SERVO) and the like. But all focusing techniques focus on solving the problem of how to make the current subject of photographing clear, rather than solving the problem of judging whether the distance of the current subject of photographing is proper. Correspondingly, the high-altitude holder camera can amplify objects reaching a range of several kilometers to a proper size by means of a powerful PTZ positioning function, and better detection, identification and tracking are realized, so that the high-altitude holder camera is greatly different from the traditional camera shooting system in focusing requirement performance.
However, existing aerial pan-tilt cameras do not have a new focusing system. But only adds some simple automatic cruising functions, and can move mechanically along some simple tracks set by people, thereby realizing the function of utilizing PTZ positioning.
Thus, the prior art has the disadvantages: the intelligent cruising device is lack of a large-range PTZ function capable of further utilizing the existing high-altitude PTZ, perceives the surrounding environment, automatically judges the distance of the main body of the visual field object, automatically realizes intelligent cruising according to the distance of the object, and achieves the function of fully utilizing the PTZ of the high-altitude PTZ to carry out intelligent cruising of the environment.
Related data: harris operator:
https://baike.baidu.com/item/Harris％20％E7％AE％97％E5％AD％90/22251459。
disclosure of Invention
In view of at least one defect in the prior art, the invention aims to provide an environment automatic sensing cruising method based on a high-altitude tripod head, which can automatically sense the distance of a main object in an image, thereby automatically realizing intelligent cruising according to the distance and the recently of the main object.
In order to achieve the above purpose, the invention adopts the following technical scheme: an environment automatic perception cruising method based on a high-altitude tripod head comprises a first stage and a second stage; firstly, obtaining an optimal Z value in the same PT value by continuously shooting pictures by utilizing the spherical view field of the existing high-altitude tripod head camera to obtain distance perception; step two, removing PTZ values corresponding to excessive redundant images, and automatically generating a proper cruising track;
stage one includes: initializing a high-altitude pan-tilt camera, and setting PTZ values of the high-altitude pan-tilt camera to be minimum values; triple-cycle scanning is carried out on the PTZ, and a photo is taken, so that an equal-ratio sequence image is formed; selecting the most suitable Z value image from the equal ratio sequence images, and storing the most suitable Z value image;
the second stage comprises the following steps: under the condition that the Z value is unchanged, removing images with redundancy exceeding the repetition proportion according to the distance between the P value and the T value; and picking out the image sequence with the nearest PZ value distance, and re-sequencing and outputting the PTZ value of the image sequence as an intelligent cruising route.
The environment automatic perception cruising method based on the high-altitude tripod head is characterized by comprising the following steps of: the first stage comprises the following steps:
step one: initializing preparation;
step 1.1: reading the range P of the P value of the high-altitude PTZ camera through the SDK or ONVIF protocol of the high-altitude PTZ camera min And P max Range of T values T min =0 and T max And a range Z of Z values min And Z max ；
Step 1.2: establishing a first Arr array and a Temp array, and initializing the first Arr array and the Temp array, wherein the first Arr array is used for storing result data of the stage one; the Temp array is used for temporarily storing the process data;
step 1.3: setting the PTZ values of the high-altitude PTZ cameras to minimum values, namely returning to P=P min ,T＝T min ,Z＝Z min State, ready to start the environment-aware shooting.
The environment automatic perception cruising method based on the high-altitude tripod head is characterized by comprising the following steps of: the first stage comprises the following steps:
step two: circularly shooting a concentric spherical image, and performing environment sensing;
triple-cycle scanning by using high-altitude pan-tilt camera, wherein the P value is from P min ～P Max Gradually changing; t value from T min ～T max Gradually change, Z value from Z min To Z max Gradually changing; the innermost loop is Z-value, increasing P-value whenever it reaches a maximum; when the P value reaches the maximum value P max At that time, the T value is increased.
The environment automatic perception cruising method based on the high-altitude tripod head is characterized by comprising the following steps of: step two comprises the following substeps:
step 2.1: in order to ensure that a proper PTZ value is obtained, the image shooting of the high-altitude tripod head camera adopts a dense strategy; let Δp=2, Δt=2, which represent the step values in the horizontal direction and the vertical direction, respectively; let Δz=1 again, indicating the increase multiple per zoom; the high altitude pan-tilt camera will be from p=p min ,T＝T min ,Z＝Z min Shooting is started;
step 2.2: judgment T>T max If yes, jumping out of the whole cycle; otherwise, continuing;
step 2.3: judging P>P max If yes, turning to step 2.9, otherwise, continuing;
step 2.4: judgment of Z>Z max If yes, turning to step 2.7, otherwise, continuing;
step 2.5: take a photograph and assemble it into objects: { "P": T ":" Z ": image data }, put it into a Temp array;
step 2.6: changing the Z value by taking deltaz as an increase multiple, and turning to the step 2.4;
step 2.7: selecting an object with a proper motion speed from the Z-value sequence image by using a Harris angular point algorithm; finding out the most suitable Z-value image from the Temp array, and putting the image into a first Arr array list;
step 2.8: resetting Z, namely: z=z min The method comprises the steps of carrying out a first treatment on the surface of the The P value increases stepwise, i.e.: adjusting the P value by taking Δp as a step value; clearing the Temp array, and turning to step 2.3;
step 2.9: reset P, i.e. p=p min The method comprises the steps of carrying out a first treatment on the surface of the The T value increases stepwise, i.e.: adjusting the T value by taking deltat as a stepping value; turning to step 2.2.
The environment automatic perception cruising method based on the high-altitude tripod head is characterized by comprising the following steps of:
outputting a first Arr array;
at this time, the first arr array is a proper Z-value image sequence corresponding to all PT values; the FirstArr array is taken as the output of stage one and also as the input of stage two.
The environment automatic perception cruising method based on the high-altitude tripod head is characterized by comprising the following steps of: step 2.7 two values are generated for each image of the Z-value sequence image using Harris corner algorithm: fastMove, slowMove; wherein FastMove represents the proportion of image blocks moving fast in the image compared with the previous image, and SlowMove represents the proportion of image blocks moving slowly; then, the image with the minimum value of FastMove+SlowMove can be selected and marked as Min (FastMove+SlowMove), and the image is regarded as the image with the right Z value; in addition, if Min (FastMove+SlowMove) >0.6 is finally selected, all subsequent images are unsuitable, only the first image can be selected, and the Z value is 1; the most suitable Z-valued image is found from the Temp array and put into the FirstArr array list.
The environment automatic perception cruising method based on the high-altitude tripod head is characterized by comprising the following steps of: the second stage comprises the following steps: initializing, wherein the initializing comprises the following steps:
step 1.10: reading a FirstArr array, and arranging and ascending the array according to a Z value, a T value and a P value;
step 1.20: establishing a SecArr array and a Result array, and initializing the SecArr array, the Result array and the Temp array to be empty.
The environment automatic perception cruising method based on the high-altitude tripod head is characterized by comprising the following steps of: the second stage further comprises a second step: removing the overlapped redundant images;
step 2.10: determine whether the FirstArr array is empty? If the image is empty, the step of removing the overlapped redundant images is finished, otherwise, the step is continued;
step 2.20: sequentially taking out an element from the first Arr array, and marking the element as E; note that all fetches contain the process of deletion from the original array.
Step 2.30: finding out the Z value same as the E element from the rest of the FirstArr array, wherein the PT distance D is smaller than D Max Is a member of the group; wherein D is Max Is a custom constant; for example, a value of 4, and put into an array Temp array; the distance D here is measured using euclidean distance; wherein F is any element in the first Arr array;
in the formula (1), D represents the distance between two elements, E p Representing the p value of element E, F p Representing the p value of element F, E t T value, F representing element E t A t value representing element F;
step 2.40: performing cyclic processing, namely performing matching operation on images of all elements S in the Temp array and images of elements E, and generating a matched Match value; examples: match=0.3, indicating that the overlap of the two images is 30%.
Step 2.50: if the value Match of the element S and the element E is greater than or equal to the Match value Match max The method comprises the steps of carrying out a first treatment on the surface of the For example, match max 50%, deleting the corresponding element of the element S from the FirstArr array; if the elements in the Temp array are not processed, turning to step 2.40; otherwise, continuing;
step 2.60: and putting the element E into an array SecArr array, and clearing a Temp array, and turning to the step 2.20.
The environment automatic perception cruising method based on the high-altitude tripod head is characterized by comprising the following steps of: the second stage further comprises a third step;
step three: further arranging the SecArr array, and outputting a final intelligent cruising route; comprising the following steps:
step 3.1: judging whether the SecArr array is empty or not, if so, turning to the step 4;
step 3.2: sequentially taking out an element E2 from the SecArr array, and putting the E2 into a Result array;
step 3.3: taking out the element Next with the same Z value as the element E2 but the shortest distance from the SecArr array, namely satisfying the following formula; wherein the distance formula uses the previous formula (1);
wherein arg minD (x, E2) represents the element x (i.e., next element) which is closest to E2; x belongs to a SecArr array; in addition, any element y in the SecArr array has a distance D (y, E2) between y and E2 greater than or equal to a distance D (x, E2) between x and E2;
step 3.4: is this element Next present? If yes, putting the element into a Result array, pointing E2 to the element Next, and turning to step 3.3; if not, turning to step 3.1;
step 4: sequentially outputting PTZ elements in the Result array, namely the found intelligent cruising route;
step 5: and (4) performing intelligent cruising according to the intelligent cruising route in the step (4).
The environment automatic perception cruising method based on the high-altitude holder has the advantages that the environment automatic perception cruising method based on the high-altitude holder can automatically perceive the distance of a main object in an image, so that intelligent cruising according to the distance and the recently of the main object is automatically realized.
Drawings
FIG. 1 is a flow chart of the method of the present invention;
FIG. 2 is a flow chart of the high altitude pan/tilt environment awareness;
FIG. 3 is a flow chart for generating an intelligent cruising path of the high altitude tripod head.
Detailed Description
The invention will be described in further detail with reference to the drawings and the specific examples.
As shown in fig. 1-3, the present invention discloses a method for performing automatic environment sensing directly using PTZ photographing information of a high altitude pan-tilt camera and performing automatic cruising using the sensing information. The high-altitude cradle head has certain intelligence, can be automatically adapted to all surrounding environments, and realizes the purpose of releasing the vector during cruising. It comprises the following steps: setting PTZ values of a high-altitude holder to be minimum values; (2) Keeping PT value unchanged, increasing Z value continuously to shoot different pictures, thereby forming an equal ratio sequence image; (3) Automatically selecting a most suitable Z value from different Z value sequence images, and storing the Z value; (4) Changing PT values, and repeating the step 2 until all PT values are traversed; (5) Under the condition that the Z value is unchanged, the distance of the value P, T is reduced, an optimal moving path is reselected by removing images (6) which are redundant and exceed the size of the repeated proportion, the stored PTZ values are output in a reordered mode, the PTZ value at the moment is the optimal PTZ value after intelligent environment sensing, cruising is carried out according to the environment according to cruising of the PTZ sequence, and intelligent cruising conditions are met. The invention can automatically sense the distance of objects in the surrounding environment by using the high-altitude holder with only a single camera, thereby ensuring that the lens can not mechanically perform focusing identification during automatic cruising, greatly improving the efficiency of automatic cruising and laying a foundation for further artificial intelligent target detection. PTZ is abbreviated as Pan/Tilt/Zoom in security monitoring application, and represents omnibearing (up-down, left-right) movement of a tripod head and Zoom control of a lens.
The invention provides a function of automatically sensing the distance of a main object in an image through the environment image shooting of a concentric sphere in advance, and then optimizing the setting of the cruising track through the sensing of the distance of the environment in advance, so as to realize the function of intelligent cruising track. The locating and identifying function of the forestry high-altitude holder can be greatly accelerated and enhanced, and the working efficiency is improved. By utilizing the method, any newly added equipment is not required to be added at the front end, and the method really realizes: "intelligent cruising with a vector".
An environment automatic perception cruising method based on a high-altitude tripod head comprises two stages: the first stage is to continuously shoot PTZ sequence images and process the PTZ sequence images so as to acquire the optimal PTZ value of a main object in the environment and further acquire distance perception, and the second stage is to further sort the acquired PTZ value images and remove the PTZ sequence with excessive image overlapping so as to realize automatic cruising:
stage one: obtaining an optimal Z value in the same PT value by continuously shooting pictures of different PTZ values by utilizing the spherical view field of the existing high-altitude tripod head camera, and obtaining distance perception;
it comprises the following steps:
step one: initialization preparation:
step 1.1: reading the camera P value range P through the camera SDK or ONVIF protocol min And P max (example: P min ＝0,P max ＝360)，P min Is the minimum value of P value, P max Is the maximum value of the P value; range of T values T min =0 and T max ,T min Is the minimum value of T max Is the maximum value of the T value; (example: T) min = -80 and T max =0), and a range Z of Z values min And Z max ；Z min Is the minimum value of Z value, Z max Is the maximum value of Z value, (for example, since Z value is 1 times minimum, there is Z min =1 to Z max =50 times).
Step 1.2: the FirstArr array is initialized, which will be used to store stage one result data, and the Temp array, which is used to temporarily store process data.
Step 1.3 returns the camera to the minimum, i.e. to p=p min ,T＝T min ,Z＝Z min State, ready to start the environment-aware shooting.
Step two: and circularly shooting the concentric spherical image to perform environment sensing.
Triple circulation of PTZScanning, wherein the P value is from P min ～P Max (0-360 degrees) and T value from T min ～T max (-90-0 degree) and Z value from Z min To Z max (in connection with the camera, e.g. 1-50 times) gradually. In order to avoid the problems of image blurring and the like caused by the movement of camera cradle head mechanical equipment, the PT value is firstly kept unchanged, the Z value is circularly changed (only the image is amplified), and the P value is increased every time the Z value reaches the maximum value; when the P value reaches the maximum value T max Or minimum value T min At that time, the T value is increased.
Step 2 comprises the following sub-steps:
step 2.1: in order to ensure that a proper PTZ value is obtained, a dense strategy is adopted for image shooting of the pan-tilt camera. Let Δp=2, Δt=2, which represent the step values in the horizontal direction and the vertical direction, respectively; let Δz=1 again, indicating the increase multiple per zoom. The camera will be from p=p min ,T＝T min ,Z＝Z min Shooting is started.
Step 2.2: judgment T>T max If yes, jumping out of the whole cycle, turning to the step 3, otherwise continuing;
step 2.3: judging P>P max If yes, turning to step 2.9, otherwise, continuing;
step 2.4: judgment of Z>Z max If yes, turning to step 2.7, otherwise, continuing;
step 2.5: take a photograph and assemble it into objects: { "P": T ":" Z ": image data }, put it into a Temp array.
Step 2.6: changing the Z value under the condition that the PT value is kept unchanged, wherein Z=Z+Δz, and turning to step 2.4;
step 2.7: because the Z value is in arithmetic increment, the nature of the shot image is the projection transformation from a three-dimensional space to a two-dimensional space, and the near-large and far-small rule is satisfied; namely: the far object moves slowly and the near object moves fast so that we can use Harris corner algorithm to generate two values for each image of the Z-value sequence image: fastMove, slowMove. Where FastMove represents the proportion of image blocks in an image that move rapidly compared to the previous image, for example: fastpove=0.2, indicating that 20% of the blocks are moving rapidly; slowMove represents the proportion of slowly moving image blocks. Then, the image with the minimum value of FastMove+SlowMove can be selected and marked as Min (FastMove+SlowMove), and the image is regarded as the image with the right Z value; in addition, if Min (FastMove+SlowMove) >0.6 is finally selected, it indicates that all subsequent images are unsuitable, and only the first image (Z value of 1) can be selected. The most appropriate Z-valued image is found from the Temp array and put into the FirstArr list.
Step 2.8: resetting Z, namely: z=z min The method comprises the steps of carrying out a first treatment on the surface of the The P value increases stepwise, i.e.: p=p+Δp. And (3) clearing the Temp array, and turning to step 2.3.
Step 2.9: reset P, i.e. p=p min The method comprises the steps of carrying out a first treatment on the surface of the The T value increases stepwise, i.e.: t=t+Δt. Turning to step 2.2.
Outputting a first Arr:
at this time, firstArr is a suitable Z-value image sequence corresponding to all PT values. FirstArr is taken as the output of stage one and also as the input of stage two.
Stage two: because the shooting sampling in the stage one adopts dense shooting, in order to accelerate intelligent cruising, PTZ values corresponding to excessive redundant images need to be removed, and a proper cruising track is automatically generated;
it comprises the following steps:
step one: initialization, which includes:
step 1.1, reading FirstArr, and arranging the FirstArr according to the Z value, the T value and the P value in ascending order.
Step 1.2 prepares to initialize the SecArr, result, temp array to null.
Step two: and removing the overlapped redundant images.
Step 2.1 determines whether FirstArr is empty? If the air is not available, the step 3 is carried out, otherwise, the process is continued.
Step 2.2 an element is sequentially fetched from FirstArr, denoted as E (note that all fetches and revenues contain the process of deletion from the original array).
Step 2.2 find the and E element from the remaining FirstArrThe Z values of the elements are the same, and the PT distance D is smaller than D Max All elements (wherein D Max Is a custom constant, e.g., 4 value) and is placed into the array Temp. The distance here is measured using euclidean distance, formula one below, where F is any element in FirstArr.
Wherein D represents the distance between two elements, E p Representing the p value of element E, F p Representing the p value of element F, and similarly E t T value, F representing element E t The t value representing element F.
Step 2.4, performing loop processing, namely performing matching operation on images of all elements S in the array Temp and images of elements E, and generating matched Match values (for example, match=0.3, and the overlapping degree of the two images is 30%).
Step 2.5 if the value Match of the Match of element S and element E is greater than or equal to Match max (e.g., 50%), the corresponding element of element S is deleted from FirstArr. If the elements in Temp are not processed, go to step 2.4. Otherwise, continuing.
Step 2.6, putting element E into an array SecArr, and emptying the Temp array, and turning to step 2.2.
And step three, further finishing the SecArr, and outputting a final intelligent cruising route. Comprising the following steps:
step 3.1, judging whether SecArr is empty or not, if so, turning to step 4;
step 3.2 sequentially take one element E2 from SecArr and put E2 into an array Result.
Step 3.3 extracts the element Next having the same Z value as the element E2 but the shortest distance from SecArr, that is, satisfies the following formula. Wherein the distance formula uses the foregoing formula (1).
Wherein arg minD (x, E2) represents taking element x, which is closest to E2; x belongs to SecArr; in SecArr, any element y has a distance D (y, E2) between y and E2 equal to or greater than a distance D (x, E2) between x and E2. The element x in the above formula is the element Next.
Step 3.4 is this element Next? If so, the element is placed in the Result array and E2 is pointed to the element Next, go to step 3.3. If not, go to step 3.1.
And step 4, outputting the PTZ elements in the Result array sequentially, namely the found intelligent cruising route.
And 5, performing intelligent cruising according to the output route of the step 4.
The method realizes the whole process of the environment automatic perception cruising mode based on the high-altitude tripod head camera, and the corresponding two flowcharts of different stages are shown in fig. 2 and 3.
Technical protection point:
an environment automatic perception cruising method based on a high-altitude tripod head is characterized by comprising the following steps: firstly, directly utilizing triple circulation of the existing PTZ (PTZ-PTZ) of the high-altitude pan-tilt camera to shoot the surrounding environment; under the condition of keeping the PT value unchanged, the Z value is increased gradually, so that shooting is performed, the judgment of the distance is realized on a single camera according to the rule of the distance of perspective projection, and the optimal Z value in the PT sequence is extracted. Next, carrying out distance judgment on the preserved PTZ sequence image, finding out the nearest image in the sequence image under the condition that the Z value is unchanged, and removing redundant adjacent images according to the size of the overlapping proportion of the images; and finally, finding out an image sequence of the latest PT value according to the condition that the Z value is unchanged, and outputting a corresponding PTZ result. So far, the high-altitude tripod head camera can intelligently cruise the surrounding environment according to the sequence.
Technical protection scope abstract:
an environment automatic perception cruising method based on a high-altitude tripod head comprises the following steps:
firstly, using the approximate invariance of the surrounding environment of the high-altitude PTZ camera and the controllability of the PTZ of the high-altitude PTZ camera to automatically sense the surrounding environment;
the second characteristic is that under the condition that the PT value is unchanged, the Z value is kept to carry out arithmetic increment (or decrement), so that an image sequence is shot;
feature three, perception stage: because the Z value is equidifferent, the nature of the shot image is the projection transformation from a three-dimensional space to a two-dimensional space, and the near-large-far-small rule is satisfied; therefore, the far object moves slowly and the near object moves quickly, so that a Harris corner algorithm can be used for selecting an object with a proper movement speed from the Z-value sequence images;
feature four, cruise route generation stage: performing de-duplication on the PTZ information image sequence in the last step, firstly finding out an image group with the PZ value distance within a certain range under the condition that the Z value is unchanged, and then deleting images with the image overlapping degree exceeding a certain threshold value from the image group;
and fifthly, under the condition that the Z value is kept unchanged, picking out an image sequence with the closest PZ value distance, and outputting the image sequence as an intelligent cruising sequence.
Finally, it should be noted that: the above description is only illustrative of the specific embodiments of the invention and it is of course possible for those skilled in the art to make modifications and variations to the invention, which are deemed to be within the scope of the invention as defined in the claims and their equivalents.
Claims (7)
1. The environment automatic perception cruising method based on the high-altitude tripod head is characterized by comprising a first stage and a second stage; firstly, obtaining an optimal Z value in the same PT value by continuously shooting pictures by utilizing the spherical view field of the existing high-altitude tripod head camera to obtain distance perception; step two, removing PTZ values corresponding to excessive redundant images, and automatically generating a proper cruising track;
stage one includes: initializing a high-altitude pan-tilt camera, and setting PTZ values of the high-altitude pan-tilt camera to be minimum values; triple cyclic scanning is carried out on the PTZ, and a photo is taken, so that an arithmetic sequence image is formed; selecting the most suitable Z value image from the arithmetic sequence images, and storing the most suitable Z value image;
the second stage comprises the following steps: under the condition that the Z value is unchanged, removing redundant images exceeding the repetition proportion according to the distance between the P value and the T value; selecting an image sequence with the closest PZ value distance, and re-sequencing and outputting PTZ values of the image sequence as an intelligent cruising route;
the first stage comprises the following steps:
step one: initializing preparation;
step 1.1: reading the range P of the P value of the high-altitude PTZ camera through the SDK or ONVIF protocol of the high-altitude PTZ camera min And P max Range of T values T min And T max And a range Z of Z values min And Z max ；
Step 1.2: initializing a first Arr array and a Temp array, wherein the first Arr array is used for storing result data of the stage one;
step 1.3: setting the PTZ values of the high-altitude PTZ cameras to minimum values, namely returning to P=P min ,T＝T min ,Z＝Z min A state in which the environment-aware photographing is ready to start;
step two: circularly shooting a concentric spherical image, and performing environment sensing;
triple-cycle scanning by using high-altitude pan-tilt camera, wherein the P value is from P min ～P Max The method comprises the steps of carrying out a first treatment on the surface of the T value from T min ～T max Z value from Z min To Z max The method comprises the steps of carrying out a first treatment on the surface of the The innermost loop is Z-value, increasing P-value whenever it reaches a maximum; when the P value reaches the maximum value P max When the T value is increased;
step two comprises the following substeps:
step 2.1: in order to ensure that a proper PTZ value is obtained, the image shooting of the high-altitude tripod head camera adopts a dense strategy; let Δp=2, Δt=2, which represent the step values in the horizontal direction and the vertical direction, respectively; let Δz=1 again, indicating the increase multiple per zoom; the high altitude pan-tilt camera will be from p=p min ,T＝T min ,Z＝Z min Shooting is started;
step 2.2: judgment T>T max If yes, jumping out of the whole cycle; otherwise, continuing;
step (a)2.3: judging P>P max If yes, turning to step 2.9, otherwise, continuing;
step 2.4: judgment of Z>Z max If yes, turning to step 2.7, otherwise, continuing;
step 2.5: take a photograph and assemble it into objects: { "P": T ":" Z ": image data }, put it into a Temp array;
step 2.6: changing the Z value by taking deltaz as an increasing step length, and turning to the step 2.4;
step 2.7: selecting an object with a proper motion speed from the Z-value sequence image by using a Harris angular point algorithm; finding out the most suitable Z-value image from the Temp array, and putting the image into a first Arr array list;
step 2.8: resetting Z, namely: z=z min The method comprises the steps of carrying out a first treatment on the surface of the The P value increases stepwise, i.e.: adjusting the P value by taking Δp as a step value; clearing the Temp array, and turning to step 2.3;
step 2.9: reset P, i.e. p=p min The method comprises the steps of carrying out a first treatment on the surface of the The T value increases stepwise, i.e.: adjusting the T value by taking deltat as a stepping value; turning to step 2.2.
2. The automatic environment sensing cruising method based on the high altitude tripod head of claim 1, wherein the method comprises the following steps:
outputting a first Arr array;
at this time, the first arr array is a proper Z-value image sequence corresponding to all PT values; the FirstArr array is taken as the output of stage one and also as the input of stage two.
3. The automatic environment sensing cruising method based on the high altitude tripod head of claim 1, wherein the method comprises the following steps: step 2.7 two values are generated for each image of the Z-value sequence image using Harris corner algorithm: fastMove, slowMove; wherein FastMove represents the proportion of image blocks moving fast in the image compared with the previous image, and SlowMove represents the proportion of image blocks moving slowly; then, the image with the minimum value of FastMove+SlowMove can be selected and marked as Min (FastMove+SlowMove), and the image is regarded as the image with the right Z value; in addition, if Min (FastMove+SlowMove) >0.6 is finally selected, all subsequent images are unsuitable, only the first image can be selected, and the Z value is 1; the most suitable Z-valued image is found from the Temp array and put into the FirstArr array list.
4. The automatic environment sensing cruising method based on the high altitude tripod head of claim 1, wherein the method comprises the following steps: the second stage comprises the following steps: initializing, wherein the initializing comprises the following steps:
step 1.10: reading a FirstArr array, and arranging and ascending the array according to a Z value, a T value and a P value;
step 1.20: the SecArr array, result array, temp array are initialized to null.
5. The automatic environment sensing cruising method based on the high altitude tripod head of claim 4, wherein the method comprises the following steps: the second stage further comprises a second step: removing the overlapped redundant images;
step 2.10: determine whether the FirstArr array is empty? If the image is empty, the step of removing the overlapped redundant images is finished, otherwise, the step is continued;
step 2.20: sequentially taking out an element from the first Arr array, and marking the element as E;
step 2.30: finding out the Z value same as the E element from the rest of the FirstArr array, wherein the PT distance D is smaller than D Max Is a member of the group; wherein D is Max Is a custom constant; placing the data into an array Temp array; the distance D here is measured using euclidean distance; wherein F is any element in the first Arr array;
in the formula (1), D represents the distance between two elements, E p Representing the p value of element E, F p Representing the p value of element F, E t T value, F representing element E t A t value representing element F;
step 2.40: performing cyclic processing, namely performing matching operation on images of all elements S in the Temp array and images of elements E, and generating a matched Match value;
step 2.50: if the value Match of the element S and the element E is greater than or equal to the Match value Match max Deleting the corresponding element of the element S from the FirstArr array; if the elements in the Temp array are not processed, turning to step 2.40; otherwise, continuing;
step 2.60: and putting the element E into an array SecArr array, and clearing a Temp array, and turning to the step 2.20.
6. The automatic environment sensing cruising method based on the high altitude tripod head of claim 5, wherein the method comprises the following steps: the second stage further comprises a third step;
step three: further arranging the SecArr array, and outputting a final intelligent cruising route; comprising the following steps:
step 3.1: judging whether the SecArr array is empty or not, if so, turning to the step 4;
step 3.2: sequentially taking out an element E2 from the SecArr array, and putting the E2 into a Result array;
step 3.3: taking out the element Next with the same Z value as the element E2 but the shortest distance from the SecArr array;
step 3.4: is this element Next present? If yes, putting the element into a Result array, pointing E2 to the element Next, and turning to step 3.3; if not, turning to step 3.1;
step 4: and outputting the PTZ elements in the Result array sequentially, namely the found intelligent cruising route.
7. The automatic environment sensing cruising method based on the high altitude tripod head of claim 6, wherein the method comprises the following steps: the second stage further comprises a step 5: and (4) performing intelligent cruising according to the intelligent cruising route in the step (4).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111185406.8A CN113905178B (en) | 2021-10-12 | 2021-10-12 | Environment automatic sensing cruising method based on high-altitude holder |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111185406.8A CN113905178B (en) | 2021-10-12 | 2021-10-12 | Environment automatic sensing cruising method based on high-altitude holder |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113905178A CN113905178A (en) | 2022-01-07 |
| CN113905178B true CN113905178B (en) | 2023-05-30 |
Family
ID=79191585
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111185406.8A Active CN113905178B (en) | 2021-10-12 | 2021-10-12 | Environment automatic sensing cruising method based on high-altitude holder |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113905178B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102231798A (en) * | 2011-06-24 | 2011-11-02 | 天津市亚安科技电子有限公司 | Method for controlling PTZ (Pan/Tilt/Zoom) camera to zoom automatically and system thereof |
| CN105338248A (en) * | 2015-11-20 | 2016-02-17 | 成都因纳伟盛科技股份有限公司 | Intelligent multi-target active tracking monitoring method and system |
| CN112180962A (en) * | 2020-09-30 | 2021-01-05 | 苏州臻迪智能科技有限公司 | Flight control method and device for unmanned aerial vehicle, electronic equipment and storage medium |
| CN112800867A (en) * | 2021-01-13 | 2021-05-14 | 重庆英卡电子有限公司 | Pine wood nematode withered tree detection method based on two-stage high-altitude pan-tilt video |
| CN112822478A (en) * | 2020-12-31 | 2021-05-18 | 杭州电子科技大学 | High-quality photo sequence acquisition method for three-dimensional reconstruction |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| AU2011265430B2 (en) * | 2011-12-21 | 2015-03-19 | Canon Kabushiki Kaisha | 3D reconstruction of partially unobserved trajectory |
| US20140078300A1 (en) * | 2012-09-14 | 2014-03-20 | Motorola Solutions, Inc. | Adjusting surveillance camera ptz tours based on historical incident data |
| AU2015234329A1 (en) * | 2015-09-30 | 2017-04-13 | Canon Kabushiki Kaisha | Method, system and apparatus for processing an image |
-
2021
- 2021-10-12 CN CN202111185406.8A patent/CN113905178B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102231798A (en) * | 2011-06-24 | 2011-11-02 | 天津市亚安科技电子有限公司 | Method for controlling PTZ (Pan/Tilt/Zoom) camera to zoom automatically and system thereof |
| CN105338248A (en) * | 2015-11-20 | 2016-02-17 | 成都因纳伟盛科技股份有限公司 | Intelligent multi-target active tracking monitoring method and system |
| CN112180962A (en) * | 2020-09-30 | 2021-01-05 | 苏州臻迪智能科技有限公司 | Flight control method and device for unmanned aerial vehicle, electronic equipment and storage medium |
| CN112822478A (en) * | 2020-12-31 | 2021-05-18 | 杭州电子科技大学 | High-quality photo sequence acquisition method for three-dimensional reconstruction |
| CN112800867A (en) * | 2021-01-13 | 2021-05-14 | 重庆英卡电子有限公司 | Pine wood nematode withered tree detection method based on two-stage high-altitude pan-tilt video |
Non-Patent Citations (1)
| Title |
|---|
| 基于时序巡航图像的茶树生长监测研究;刘连忠;李孟杰;宁井铭;;浙江农业学报(第05期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113905178A (en) | 2022-01-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11765339B2 (en) | Estimating pose in 3D space | |
| CN113038016B (en) | Unmanned aerial vehicle image acquisition method and unmanned aerial vehicle | |
| CN110663246B (en) | Method and system for processing images | |
| CN110799921A (en) | Shooting method and device and unmanned aerial vehicle | |
| JP4237212B2 (en) | Image processing method, image processing program, and program recording medium | |
| CN112710318A (en) | Map generation method, route planning method, electronic device, and storage medium | |
| CN111765974B (en) | Wild animal observation system and method based on miniature refrigeration thermal infrared imager | |
| CN108513642B (en) | Image processing method, unmanned aerial vehicle, ground console and image processing system thereof | |
| CN109765932A (en) | A kind of desert shrubbery cover degree unmanned plane investigation method | |
| CN112207821A (en) | Target searching method of visual robot and robot | |
| CN113391644B (en) | Unmanned aerial vehicle shooting distance semi-automatic optimization method based on image information entropy | |
| CN112419363B (en) | Unmanned aerial vehicle target following method and system | |
| CN113905178B (en) | Environment automatic sensing cruising method based on high-altitude holder | |
| CN112422905B (en) | Image acquisition method, device, equipment and medium for power equipment | |
| CN110036411B (en) | Apparatus and method for generating electronic three-dimensional roaming environment | |
| CN113597754A (en) | Method and device for acquiring match picture and method and device for controlling shooting device | |
| CN112585945A (en) | Focusing method, device and equipment | |
| CN115550549A (en) | Intelligent region of interest eagle eye observation method based on unmanned aerial vehicle | |
| CN112640420B (en) | Control method, device, equipment and system of electronic device | |
| CN111684784B (en) | Image processing method and device | |
| CN114697528A (en) | Image processor, electronic device and focusing control method | |
| CN112585946A (en) | Image shooting method, image shooting device, movable platform and storage medium | |
| JP6260803B2 (en) | Imaging system, moving body, imaging method, and program | |
| CN112154477A (en) | Image processing method and device and movable platform | |
| JP2019185541A (en) | Image processing apparatus, image processing method, and program |
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 |