CN102701056A - Dynamic image recognition control method, device and system - Google Patents
Dynamic image recognition control method, device and system Download PDFInfo
- Publication number
- CN102701056A CN102701056A CN2011104102354A CN201110410235A CN102701056A CN 102701056 A CN102701056 A CN 102701056A CN 2011104102354 A CN2011104102354 A CN 2011104102354A CN 201110410235 A CN201110410235 A CN 201110410235A CN 102701056 A CN102701056 A CN 102701056A
- Authority
- CN
- China
- Prior art keywords
- dynamic image
- moving target
- unit
- elevator door
- current dynamic
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Indicating And Signalling Devices For Elevators (AREA)
Abstract
The invention discloses a dynamic image recognition control method, which comprises the following steps of: acquiring a dynamic image in an image data register; extracting a moving target in the current dynamic image; and tracking the moving target and judging whether the moving target has the tendency of moving toward an elevator door, if so, generating control information of opening the elevator door. The invention also discloses a dynamic image recognition control device and a dynamic image recognition control device. The dynamic image recognition control method is easy to operate and convenient to use, any external equipment is not required, a dynamic image in a certain area outside the elevator sedan can be detected, the moving target in the dynamic image is extracted, analyzed and tracked, and whether the moving target has the tendency of moving toward the elevator door is judged according to the motion trail of the moving target so as to control delayed closing or reopening of the elevator door, so that the opening and closing of the elevator door are intelligent, and the user requirements are met.
Description
Technical field
The present invention relates to field of intelligent control, relate in particular to a kind of dynamic image identification control method, Apparatus and system.
Background technology
The basis control of elevator comprises that manipulation refers to that layer case and outer-office calling refer to a layer case in the sedan-chair.Handle in the sedan-chair and refer to that a layer case is used for sending instruction task in the sedan-chair, the order elevator starter with stop layer and pull in and show elevator service direction and position; Outer-office calling refers to that layer case is the device of the calling elevator that provides for the personnel that outside the Room, prepare using escalator, and it has the uplink and downlink button, can also show position and service direction that elevator is present.Generally have proximity protection device in addition on the elevator door, elevator door safety edge drive interlock switch or opto-electronic pickup, the nearly door sensor master cock of induction type door motor prevent elevator folder people, folder thing.But elevator all needs manually press in the car and accomplishes with the outer control button in the Room switch activated or when stopping layer operation, and elevator can not find voluntarily that the someone has the using escalator intention, make the passenger often because of the difference in several steps by the pass outside elevator.
At present, begun both at home and abroad above-mentioned defective is carried out correlative study.
Like one Chinese patent application CN90101587.3; Denomination of invention is in " apparatus for controlling elevator "; Mention this apparatus for controlling elevator and use the situation control car up-down that waits the passenger in visitor and the car of the elevator waiting space that obtains by camera head; But the static distribution situation that this only relates to the inside and outside passenger of lift car does not relate to the judgement to passenger moving trend.
One Chinese patent application CN02828956.0 and for example; Denomination of invention is in " encumbrance in the elevator door and detect towards the neural network that moves of elevator door "; Mention the pick up camera that utilization can suitably throw light on image is offered transaction card, it is transformed into image value of vectors and value of vectors is offered neural network.This neural network can according to object in the path of elevator door or object on floor platform, move with the mode of wanting to get into elevator near elevator, thereby the enabling signal of providing.This enabling signal is offered the elevator door control device,, make this door open or stay open so that be to move or object is arranged in this elevator door access road towards elevator according to object.But this system also need add computing machine before detecting learns with other image processing software, is unfavorable for user's operation.
Summary of the invention
Technical matters to be solved by this invention is; A kind of easy to operate dynamic image identification control method, Apparatus and system are provided; Can detect movement tendency to the dynamic image in the outer certain zone of lift car, thereby realize the switch of elevator door is controlled with judgement people or object.
In order to solve the problems of the technologies described above; The invention provides a kind of dynamic image identification control method; Comprise: obtain the dynamic image in the image data buffer; Wherein, the said dynamic image of gathering in real time is made as current dynamic image, the last dynamic image of said current dynamic image is made as preposition dynamic image; Extract the moving target in the said current dynamic image; Follow the tracks of said moving target and judge whether said moving target has towards the trend of elevator door motion, is judged as when being, generates the control information of opening elevator door.
As the improvement of such scheme, the step of the moving target in the current dynamic image of said extraction comprises: judge foreground object according to said current dynamic image and background image; Generate the number of moving target and definite said moving target according to said foreground object.
As the improvement of such scheme, saidly generate moving target and confirm that the step of the number of moving target comprises: said foreground object is carried out binary conversion treatment and judges connect domain according to foreground object; Generate the number of moving target and definite moving target according to the relation of the position between said connect domain.
As the improvement of such scheme, the step of the moving target in the current dynamic image of said extraction also comprises: said current dynamic image is carried out brightness normalization method handle.
Improvement as such scheme; Said pursuit movement target is also judged whether moving target has towards the step of the trend of elevator door motion and comprised: obtain the lead position of said moving target, said lead position is the real time position that said moving target is write down in preposition dynamic image; Lead position according to said moving target generates the estimated position; Said estimated position and moving target are mated, confirm the real time position of said moving target; Generate path of motion according to the real time position in said lead position and the current dynamic image; Judge according to said path of motion whether said moving target has towards the trend of elevator door motion, is judged as when being, said moving target is the trend that has towards the elevator door motion.
As the improvement of such scheme, the generation step of said background image comprises: obtain current dynamic image and preposition dynamic image in the image data buffer; Judge in the said current dynamic image whether have mobile according to said current dynamic image and preposition dynamic image,, said current dynamic image is made as background image when the judgement of continuous said current dynamic image all is not the time.
Correspondingly; The present invention also provides a kind of dynamic image identification control device; Comprise: the acquiring unit that is used for obtaining the dynamic image of image data buffer; Wherein, the said dynamic image of gathering in real time is made as current dynamic image, the last dynamic image of said current dynamic image is made as preposition dynamic image; Link to each other with said acquiring unit, be used for extracting the extraction unit of the moving target of said current dynamic image; Link to each other with said extraction unit, be used to follow the tracks of said moving target and judge whether said moving target has towards the judging unit of the trend of elevator door motion; Link to each other with said judging unit, be used for when said judgment unit judges goes out said moving target and has the trend towards elevator door motion, generating the control unit of the control information of opening elevator door.
As the improvement of such scheme, said extraction unit comprises: be used for the prospect judging unit according to said current dynamic image and background image judgement foreground object; Link to each other with said prospect judging unit, be used for generating the target generation unit of the number of moving target and definite said moving target according to said foreground object.
As the improvement of such scheme, said target generation unit comprises: be used for said foreground object is carried out binary conversion treatment and judges the Region Segmentation subelement of connect domain; Link to each other with said Region Segmentation subelement, be used for generating the Target Fusion subelement of the number of moving target and definite moving target according to the relation of the position between said connect domain.
As the improvement of such scheme, said extraction unit also comprises and is used for said current dynamic image is carried out the brightness normalization method unit that brightness normalization method is handled.
As the improvement of such scheme, said extraction unit also comprises: link to each other with said brightness normalization method unit, be used for judging according to said current dynamic image and preposition dynamic image whether said current dynamic image exists the judgement unit of mobile; Link to each other with said judgement unit and the prospect judging unit links to each other respectively, be used for when said judgement unit is judged continuous said current dynamic image and do not had mobile, said current dynamic image being made as the background generation unit of background image.
As the improvement of such scheme, said judging unit comprises: be used to obtain the lead position acquiring unit of the lead position of said moving target, said lead position is the real time position that said moving target is write down in preposition dynamic image; Link to each other with said lead position acquiring unit, be used for generating the estimated position generation unit of estimated position according to the lead position of said moving target; Link to each other with said estimated position generation unit, be used for said estimated position and moving target mated with the real time position of the real time position of confirming said moving target and confirm the unit; Link to each other with said position determination unit, be used for generating the track generation unit of path of motion according to the real time position of said lead position and current dynamic image; Link to each other with said track generation unit, be used for judging according to said path of motion whether said moving target has towards the identifying unit of the trend of elevator door motion.
Correspondingly, the present invention also provides a kind of dynamic image control system for identifying, comprising: the imageing sensor that is used to gather dynamic image; Be used to store the image data buffer of said dynamic image; Be used for extracting moving target and moving target being analyzed tracking, judge according to the path of motion of moving target whether moving target has the dynamic image identification control device of trend to control the elevator door late release or to open again towards the elevator door motion according to said dynamic image.
The beneficial effect of embodiment of the present invention is: the dynamic image to the outer certain zone of lift car detects; Extract the moving target in the dynamic image and moving target analyzed tracking, judge according to the path of motion of moving target whether moving target has the trend of moving towards elevator door.When judging moving target is when the intention that gets into elevator is arranged, and control elevator door late release or open again makes the switch of elevator door intelligent more.In addition, need not in the course of the work to take over what external device, simple to operate, easy to use, satisfy user's demand.
Description of drawings
Fig. 1 is the first embodiment diagram of circuit of a kind of dynamic image identification control of the present invention method;
Fig. 2 is the second embodiment diagram of circuit of a kind of dynamic image identification control of the present invention method;
Fig. 3 is the 3rd an embodiment diagram of circuit of a kind of dynamic image identification control of the present invention method;
Fig. 4 is the 4th an embodiment diagram of circuit of a kind of dynamic image identification control of the present invention method;
Fig. 5 is the structural representation of a kind of dynamic image control system for identifying of the present invention;
Fig. 6 is the structural representation of dynamic image identification control device 3 in a kind of dynamic image control system for identifying of the present invention;
Fig. 7 is the structural representation of extraction unit 32 in the dynamic image identification control device 3;
Fig. 8 is another structural representation of extraction unit 32 in the dynamic image identification control device 3;
Fig. 9 is the structural representation of judging unit 33 in the dynamic image identification control device 3.
The specific embodiment
For making the object of the invention, technical scheme and advantage clearer, will combine accompanying drawing that the present invention is done to describe in detail further below.
Fig. 1 is the first embodiment diagram of circuit of dynamic image identification control method, comprising:
101, obtain dynamic image in the image data buffer.
Need to prove; Said dynamic image is stored in the said image data buffer with the form of frame successively; Wherein, the said dynamic image of gathering in real time is made as current dynamic image, the last dynamic image of said current dynamic image is made as preposition dynamic image.
102, the moving target in the said current dynamic image of extraction.
Include moving target and background image in the said current dynamic image; Wherein, said background image is the short-term unmodified, and moving target is a real time kinematics; Therefore can moving target be separated with background image, extract the moving target in the said current dynamic image effectively.
103, follow the tracks of said moving target and judge whether said moving target has towards the trend of elevator door motion.
After extracting the moving target in the said current dynamic image, need carry out tracking processing to said moving target.Because the path of motion of moving target is dynamic change, path of motion is complicated and changeable, and has only when moving target has the trend towards the elevator door motion, and elevator door just can be opened or time-delay closing again.
Need to prove that Kalman filtering capable of using is followed the tracks of Moving Target.
104, be judged as when being, generate the control information of opening elevator door.
When judging moving target and have the trend towards elevator door motion, the control information that generates the opening elevator door is opened or time-delay closing with the control elevator door again, makes the switch of elevator door realize intelligent.
Fig. 2 is the second embodiment diagram of circuit of dynamic image identification control method, comprising:
201, obtain dynamic image in the image data buffer.
Wherein, the said dynamic image of gathering in real time is made as current dynamic image, the last dynamic image of said current dynamic image is made as preposition dynamic image.
202, judge foreground object according to said current dynamic image and background image.
In the working process,,, promptly be judged as the pixel of foreground object when the absolute value of the difference that obtains during greater than presetted pixel with cooresponding pixel in the pixel subtracting background image of current dynamic image.
Need to prove that include foreground object and background image in the said current dynamic image, wherein, said background image is the short-term unmodified, and foreground object can be people or moving articles, is real time kinematics.Therefore can the pixel of current dynamic image and the pixel of background image be compared, extract the foreground object in the said current dynamic image effectively.
203, generate the number of moving target and definite said moving target according to said foreground object.
Need to prove; Comprise one or more moving targets in the foreground object; Because separate between moving target, be independent of each other, therefore need the moving target independent separate with the number of confirming moving target to analyze the movement tendency of all moving target self-movements.
204, obtain the lead position of said moving target.
Need to prove, the said dynamic image of gathering in real time is made as current dynamic image, the last dynamic image of said current dynamic image is made as preposition dynamic image.Wherein, said lead position is the real time position that said moving target is write down in preposition dynamic image.
205, the lead position according to said moving target generates the estimated position.
206, said estimated position and moving target are mated, confirm the real time position of said moving target.
207, generate path of motion according to the real time position in said lead position and the current dynamic image.
Estimate the current estimated position of said moving target according to the lead position of moving target, the moving target in said estimated position and the current dynamic image is mated, thereby confirm the real time position of moving target, realize tracking said moving target.
208, judge according to said path of motion whether said moving target has towards the trend of elevator door motion.
When the real time position of moving target during more near elevator door, representes then that mobile has towards the trend of elevator door motion than lead position.
209, be judged as when being, generate the control information of opening elevator door.
When the real time position of moving target during more near elevator door, representes then that mobile has towards the trend of elevator door motion than lead position.Correspondingly, should generate the open and-shut mode of the control information of opening elevator door with the control elevator door.
Need to prove that the quantity of moving target is arbitrarily,, promptly need generate the control information of opening elevator door as long as there is arbitrary moving target to have towards the trend of elevator door motion.
For example, sometime, object A, object B and object C appear at respectively in the surveyed area, and wherein object A, object B move to the elevator door direction, and object C elevator door direction motion dorsad.After obtaining the dynamic image that is stored in the image data buffer, with cooresponding pixel in the pixel subtracting background image of current dynamic image to confirm foreground object.Generate moving target A, moving target B and moving target C according to said foreground object, simultaneously, generate the estimated position respectively according to the lead position of moving target A, moving target B and moving target C.The estimated position of moving target A and three moving target A, B, C in the current dynamic image are mated, judge the real time position of moving target A and form the path of motion of moving target A, judge moving target A and have trend to the elevator door motion; The estimated position of moving target B and three moving target A, B, C in the current dynamic image are mated, judge the real time position of moving target B and form the path of motion of moving target B, judge moving target B and have trend to the elevator door motion; The estimated position of moving target C and three moving target A, B, C in the current dynamic image are mated, judge the real time position of moving target C and form the path of motion of moving target C, judge moving target C not to the trend of elevator door motion.Comprehensive judged result can know that moving target A and moving target B have the trend to the elevator door motion, therefore generates the control information of opening elevator door.
Fig. 3 is the 3rd an embodiment diagram of circuit of dynamic image identification control method, comprising:
301, obtain dynamic image in the image data buffer.
302, said current dynamic image being carried out brightness normalization method handles.
Need to prove; The current dynamic image of storing in the said image data buffer is a gray level image; Therefore, need current dynamic image be carried out the normalized processing of brightness, with the various piece equalization of current dynamic image; Normalizing makes Illumination intensity not influence later judgement to identical luminance state.
303, judge foreground object according to said current dynamic image and background image.
304, said foreground object is carried out binary conversion treatment and judges connect domain.
Through binary conversion treatment, foreground object is demonstrated tangible black and white effect, highlight the profile of target effectively, foreground object is divided into a plurality of disjunct independently connect domains.
305, generate the number of moving target and definite moving target according to the relation of the position between said connect domain.
In the course of the work, carry out Target Fusion according to the position between said connect domain relation, when two independently the distance of connect domain is less than predeterminable range, judge these two independently connect domain belong to same moving target; When two independently the distance of connect domain is more than or equal to predeterminable range, judge these two independently connect domain be two independent motion targets.After generating moving target, confirm the number of moving target.
For example, a people's clothes belongs to different connect domains with hair because color is different, but as long as the distance of these two connect domains just can judge that less than predeterminable range these two connect domains belong to same moving target.
306, obtain the lead position of said moving target.
Need to prove that said lead position is the real time position that said moving target is write down in preposition dynamic image.
307, the lead position according to said moving target generates the estimated position;
308, said estimated position and moving target are mated, confirm the real time position of said moving target.
309, generate path of motion according to the real time position in said lead position and the current dynamic image.
Estimate the current estimated position of said moving target according to the lead position of moving target, the moving target in said estimated position and the current dynamic image is mated, thereby confirm the real time position of moving target, realize tracking said moving target.
310, judge according to said path of motion whether said moving target has towards the trend of elevator door motion.
When the real time position of moving target during more near elevator door, representes then that mobile has towards the trend of elevator door motion than lead position.
311, be judged as when being, generate the control information of opening elevator door.
For example, sometime, object A, object B and object C appear at respectively in the surveyed area, and wherein object A moves to the elevator door direction, object B transfixion, and object C elevator door direction motion dorsad.After obtaining the dynamic image that is stored in the image data buffer, said current dynamic image is carried out brightness normalization method handle the various piece equalization that makes current dynamic image.Will be in the pixel subtracting background image of the current dynamic image after brightness normalization method is handled cooresponding pixel to confirm foreground object.With foreground object through binary conversion treatment to demonstrate tangible black and white effect, highlight the profile of target effectively, foreground object is divided into a plurality of disjunct independently connect domains.Said independently connect domain is carried out analyzing and processing in twos, when two independently the distance of connect domain is less than predeterminable range, judge these two independently connect domain belong to same moving target; When two independently the distance of connect domain is more than or equal to predeterminable range, judge these two independently connect domain be two independent motion targets.After this, said independently connect domain is fused into three independent motion targets, is respectively moving target A, moving target B and moving target C.Simultaneously, the lead position according to moving target A, moving target B and moving target C generates the estimated position respectively.The estimated position of moving target A and three moving target A, B, C in the present image are mated; Match the real time position of moving target A, judge moving target A according to the path of motion of moving target A from the lead position to the real time position again and have trend to the elevator door motion; The estimated position of moving target B and three moving target A, B, C in the present image are mated, match the real time position of moving target B, judge the trend that moving target B does not move to elevator door according to moving target B path of motion; The estimated position of moving target C and three moving target A, B, C in the present image are mated, find the real time position of moving target C,, judge the trend that moving target C does not move to elevator door according to the path of motion of moving target C.Comprehensive judged result can know that moving target A has the trend to the elevator door motion, therefore generates the control information of opening elevator door.
Fig. 4 is the 4th an embodiment diagram of circuit of dynamic image identification control method, and the generation step of said background image comprises:
401, obtain current dynamic image and preposition dynamic image in the image data buffer.
Need to prove; Said dynamic image is stored in the said image data buffer with the form of frame successively; Wherein, the said dynamic image of gathering in real time is made as current dynamic image, the last dynamic image of said current dynamic image is made as preposition dynamic image.
402, judge in the said current dynamic image whether have mobile according to said current dynamic image and preposition dynamic image.
403, when the judgement of continuous said current dynamic image all be not the time, said current dynamic image is made as background image.
Deduct the pixel of preposition dynamic image with the pixel of current dynamic image,, then represent to have moving target in the current dynamic image when difference between dynamic image during greater than predetermined threshold value; When there is not mobile in the current dynamic image of continuous some, represent that then current dynamic image for there not being the background of moving target, promptly is made as background image to current dynamic image.
Need to prove that said background image is a real-time update.
Fig. 5 is the structural representation of dynamic image control system for identifying, comprising:
Be used to gather the imageing sensor 1 of dynamic image.
Need to prove that said imageing sensor 1 carries out white balance and smoothing processing for the dynamic image that obtains, wherein, white balance is handled and can be improved the colour cast phenomenon that produces because of light source when taking, and compensates through strengthening complementary color.In addition, smoothing processing can make the mild gradual change of brightness of dynamic video image effectively, reduces picture noise, improves the quality of dynamic image.
Be used to store the image data buffer 2 of said dynamic image.
Link to each other with said imageing sensor 1; Be used for extracting moving target and moving target being analyzed tracking, judge according to the path of motion of moving target whether moving target has the dynamic image identification control device 3 of trend to control the elevator door late release or to open again towards the elevator door motion according to said dynamic image.
Need to prove that the dynamic image that said imageing sensor 1 is gathered is obtained and is stored in the image data buffer 2 by said dynamic image identification control device 3.Correspondingly, when dynamic image identification control device 3 calls dynamic image, need in image data buffer 2, obtain.
In the working process; Dynamic image identification control device 3 obtains the dynamic image that is stored in the image data buffer 2; Judge the movement tendency of moving target in the surveyed area and generate control information; When judging moving target and have the trend towards elevator door motion, the control information that generates the opening elevator door is opened or time-delay closing with the control elevator door again, makes the switch of elevator door intelligent more.In addition, image data buffer 2 stores said dynamic image to make things convenient for dynamic image identification control device 3 real-time callings.
Fig. 6 is the structural representation of dynamic image identification control device 3 in the dynamic image control system for identifying, comprising:
Be used for obtaining the acquiring unit 31 of the dynamic image of image data buffer 2.
Wherein, the said dynamic image of gathering in real time is made as current dynamic image, the last dynamic image of said current dynamic image is made as preposition dynamic image.
Link to each other with said acquiring unit 31, be used for extracting the extraction unit 32 of the moving target of said current dynamic image.
Link to each other with said extraction unit 32, be used to follow the tracks of said moving target and judge whether said moving target has towards the judging unit 33 of the trend of elevator door motion.
Link to each other with said judging unit 33, be used for when said judging unit 33 is judged said moving target and had the trend towards elevator door motion, generating the control unit 34 of the control information of opening elevator door.
Dynamic image identification control device 3 obtains the dynamic image that is stored in the image data buffer 2 through acquiring unit 31.Extraction unit 32 judges according to the moving target in the said current dynamic image of extraction in the said current dynamic image and through judging unit 33 whether said moving target has towards the trend of elevator door motion; When judging moving target and have the trend towards elevator door motion, the control information that is generated the opening elevator doors by control unit 34 is opened or time-delay closing with the control elevator door again.
Fig. 7 is the structural representation of extraction unit 32 in the dynamic image identification control device 3, and said extraction unit 32 comprises:
Be used for prospect judging unit 321 according to said current dynamic image and background image judgement foreground object.
Need to prove; Be preset with presetted pixel in the prospect judging unit 321; Prospect judging unit 321 is cooresponding pixel in the pixel subtracting background image of current dynamic image, when the absolute value of the difference that obtains during greater than presetted pixel, promptly is judged as the pixel of foreground object.Include foreground object and background image in the said current dynamic image, wherein, said background image is the short-term unmodified, and foreground object can be people or moving articles, is real time kinematics.Therefore can the pixel of preceding dynamic image and the pixel of background image be compared, extract the foreground object in the said current dynamic image effectively.
Link to each other with said prospect judging unit 321, be used for generating the target generation unit 322 of the number of moving target and definite said moving target according to said foreground object.
Need to prove; Comprise one or more moving targets in the foreground object; Because separate between moving target, be independent of each other, therefore need through target generation unit 322 the moving target independent separate with the number of confirming moving target to analyze the movement tendency of all moving target self-movements.
Wherein, said target generation unit 322 comprises:
Be used for said foreground object is carried out binary conversion treatment and judges the Region Segmentation subelement 3221 of connect domain.Through binary conversion treatment, foreground object is demonstrated tangible black and white effect, highlight the profile of target effectively, foreground object is divided into a plurality of disjunct independently connect domains.
Link to each other with said Region Segmentation subelement 3221, be used for generating the Target Fusion subelement 3222 of the number of moving target and definite moving target according to the relation of the position between said connect domain.
Be provided with predeterminable range in the Target Fusion subelement 3222; In the course of the work; Target Fusion subelement 3222 carries out Target Fusion according to the relation of the position between said connect domain; When two independently the distance of connect domain is less than predeterminable range, judge these two independently connect domain belong to same moving target; When two independently the distance of connect domain is more than or equal to predeterminable range, judge these two independently connect domain be two independent motion targets.After generating moving target, confirm the number of moving target.
Fig. 8 is another structural representation of extraction unit 32 in the dynamic image identification control device 3, and is as shown in the figure, and said extraction unit 32 also comprises and is used for said current dynamic image is carried out the brightness normalization method unit 323 that brightness normalization method is handled.
Need to prove; The current dynamic image of storage is a gray level image in the said image data buffer 2; Therefore, need current dynamic image be carried out the normalized processing of brightness through brightness normalization method unit 323, with the various piece equalization of current dynamic image; Normalizing makes Illumination intensity not influence later judgement to identical luminance state.
More preferably, said extraction unit 32 also comprises:
Link to each other with said brightness normalization method unit 323, be used for judging according to said current dynamic image and preposition dynamic image whether said current dynamic image exists the judgement unit 324 of mobile.
Link to each other with said judgement unit 324 and prospect judging unit 321 links to each other respectively; Be used for when said judgement unit is judged continuous said current dynamic image and do not had mobile, said current dynamic image being made as the background generation unit 325 of background image.
Need to prove; Said dynamic image is stored in the said image data buffer with the form of frame successively; Wherein, the said dynamic image of gathering in real time is made as current dynamic image, the last dynamic image of said current dynamic image is made as preposition dynamic image.
Be provided with pre-set threshold value in the judgement unit 324, the pixel of the current dynamic image of judgement unit 324 usefulness deducts the pixel of preposition dynamic image, when difference between dynamic image during greater than predetermined threshold value, then representes to have moving target in the current dynamic image; When there is not mobile in the current dynamic image of continuous some, represent that then current dynamic image for there not being the background of moving target, promptly is made as background image to current dynamic image.
Fig. 9 is the structural representation of judging unit 33 in the dynamic image identification control device 3, and is as shown in the figure, and said judging unit 33 comprises:
Be used to obtain the lead position acquiring unit 331 of the lead position of said moving target.
Said lead position is the real time position that said moving target is write down in preposition dynamic image.
Link to each other with said lead position acquiring unit 331, be used for generating the estimated position generation unit 332 of estimated position according to the lead position of said moving target.
Link to each other with said estimated position generation unit 332, be used for said estimated position and moving target mated with the real time position of the real time position of confirming said moving target and confirm unit 333;
Link to each other with said position determination unit 333, be used for generating the track generation unit 334 of path of motion according to the real time position of said lead position and current dynamic image;
Link to each other with said track generation unit 334, be used for judging according to said path of motion whether said moving target has towards the identifying unit 335 of the trend of elevator door motion.
Need to prove, the said dynamic image of gathering in real time is made as current dynamic image, the last dynamic image of said current dynamic image is made as preposition dynamic image.
Lead position acquiring unit 331 obtains the lead position of the preposition image of said moving target.Estimated position generation unit 332 estimates the current estimated position of said moving target according to the lead position of moving target.Position determination unit 333 matees said estimated position and moving target to confirm the real time position of said moving target; Track generation unit 334 generates path of motion according to the real time position in said lead position and the current dynamic image, thereby realizes the tracking to said moving target.The real time position of judging moving target according to path of motion when identifying unit 335 during more near elevator door, representes then that mobile has towards the trend of elevator door motion than lead position.Correspondingly, should generate the open and-shut mode of the control information of opening elevator door with the control elevator door.
Need to prove that the quantity of moving target is arbitrarily,, promptly need generate the control information of opening elevator door as long as there is arbitrary moving target to have towards the trend of elevator door motion.
By on can know; Adopt the present invention can the dynamic image in the outer certain zone of lift car be detected; Extract the moving target in the dynamic image and moving target analyzed tracking, judge according to the path of motion of moving target whether moving target has the trend of moving towards elevator door.When judging moving target and have the intention that gets into elevator, control elevator door late release or open again makes the switch of elevator door intelligent more.In addition, need not in the course of the work to take over what external device, simple to operate, easy to use, satisfy user's demand.
The above is a preferred implementation of the present invention; Should be pointed out that for those skilled in the art, under the prerequisite that does not break away from the principle of the invention; Can also make some improvement and retouching, these improvement and retouching also are regarded as protection scope of the present invention.
Claims (13)
1. a dynamic image identification control method is characterized in that, comprising:
Obtain the dynamic image in the image data buffer, wherein, the said dynamic image of gathering in real time is made as current dynamic image, the last dynamic image of said current dynamic image is made as preposition dynamic image;
Extract the moving target in the said current dynamic image;
Follow the tracks of said moving target and judge whether said moving target has towards the trend of elevator door motion, is judged as when being, generates the control information of opening elevator door.
2. dynamic image identification control method as claimed in claim 1 is characterized in that, the step of the moving target in the current dynamic image of said extraction comprises:
Judge foreground object according to said current dynamic image and background image;
Generate the number of moving target and definite said moving target according to said foreground object.
3. dynamic image identification control method as claimed in claim 2 is characterized in that, saidly generates moving target and confirms that the step of the number of moving target comprises according to foreground object:
Said foreground object is carried out binary conversion treatment and judges connect domain;
Generate the number of moving target and definite moving target according to the relation of the position between said connect domain.
4. like claim 2 or 3 described dynamic image identification control methods, it is characterized in that the step of the moving target in the current dynamic image of said extraction also comprises:
Said current dynamic image is carried out brightness normalization method to be handled.
5. dynamic image identification control method as claimed in claim 1 is characterized in that, said pursuit movement target is also judged whether moving target has towards the step of the trend of elevator door motion and comprised:
Obtain the lead position of said moving target, said lead position is the real time position that said moving target is write down in preposition dynamic image;
Lead position according to said moving target generates the estimated position;
Said estimated position and moving target are mated, confirm the real time position of said moving target;
Generate path of motion according to the real time position in said lead position and the current dynamic image;
Judge according to said path of motion whether said moving target has towards the trend of elevator door motion, is judged as when being, said moving target is the trend that has towards the elevator door motion.
6. dynamic image identification control method as claimed in claim 2 is characterized in that, the generation step of said background image comprises:
Obtain current dynamic image and preposition dynamic image in the image data buffer;
Judge in the said current dynamic image whether have mobile according to said current dynamic image and preposition dynamic image,, said current dynamic image is made as background image when the judgement of continuous said current dynamic image all is not the time.
7. a dynamic image identification control device is characterized in that, comprising:
Be used for obtaining the acquiring unit of the dynamic image of image data buffer, wherein, the said dynamic image of gathering in real time be made as current dynamic image, the last dynamic image of said current dynamic image is made as preposition dynamic image;
Link to each other with said acquiring unit, be used for extracting the extraction unit of the moving target of said current dynamic image;
Link to each other with said extraction unit, be used to follow the tracks of said moving target and judge whether said moving target has towards the judging unit of the trend of elevator door motion;
Link to each other with said judging unit, be used for when said judgment unit judges goes out said moving target and has the trend towards elevator door motion, generating the control unit of the control information of opening elevator door.
8. dynamic image identification control device as claimed in claim 7 is characterized in that, said extraction unit comprises:
Be used for prospect judging unit according to said current dynamic image and background image judgement foreground object;
Link to each other with said prospect judging unit, be used for generating the target generation unit of the number of moving target and definite said moving target according to said foreground object.
9. dynamic image identification control device as claimed in claim 8 is characterized in that, said target generation unit comprises:
Be used for said foreground object is carried out binary conversion treatment and judges the Region Segmentation subelement of connect domain;
Link to each other with said Region Segmentation subelement, be used for generating the Target Fusion subelement of the number of moving target and definite moving target according to the relation of the position between said connect domain.
10. dynamic image identification control device as claimed in claim 8 is characterized in that, said extraction unit also comprises and is used for said current dynamic image is carried out the brightness normalization method unit that brightness normalization method is handled.
11. dynamic image identification control device as claimed in claim 8 is characterized in that, said extraction unit also comprises:
Link to each other with said brightness normalization method unit, be used for judging according to said current dynamic image and preposition dynamic image whether said current dynamic image exists the judgement unit of mobile;
Link to each other with said judgement unit and the prospect judging unit links to each other respectively, be used for when said judgement unit is judged continuous said current dynamic image and do not had mobile, said current dynamic image being made as the background generation unit of background image.
12. dynamic image identification control device as claimed in claim 7 is characterized in that, said judging unit comprises:
Be used to obtain the lead position acquiring unit of the lead position of said moving target, said lead position is the real time position that said moving target is write down in preposition dynamic image;
Link to each other with said lead position acquiring unit, be used for generating the estimated position generation unit of estimated position according to the lead position of said moving target;
Link to each other with said estimated position generation unit, be used for said estimated position and moving target mated with the real time position of the real time position of confirming said moving target and confirm the unit;
Link to each other with said position determination unit, be used for generating the track generation unit of path of motion according to the real time position of said lead position and current dynamic image;
Link to each other with said track generation unit, be used for judging according to said path of motion whether said moving target has towards the identifying unit of the trend of elevator door motion.
13. a dynamic image control system for identifying is characterized in that, comprising:
Be used to gather the imageing sensor of dynamic image;
Be used to store the image data buffer of said dynamic image;
Like each described dynamic image identification control device of claim 7 ~ 12.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011104102354A CN102701056A (en) | 2011-12-12 | 2011-12-12 | Dynamic image recognition control method, device and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011104102354A CN102701056A (en) | 2011-12-12 | 2011-12-12 | Dynamic image recognition control method, device and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102701056A true CN102701056A (en) | 2012-10-03 |
Family
ID=46894199
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2011104102354A Pending CN102701056A (en) | 2011-12-12 | 2011-12-12 | Dynamic image recognition control method, device and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102701056A (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108529367A (en) * | 2017-11-21 | 2018-09-14 | 江苏鸿山鸿锦物联技术有限公司 | A kind of elevator door state patch detection method based on video image |
| CN108529363A (en) * | 2018-04-26 | 2018-09-14 | 珠海亿智电子科技有限公司 | A kind of intelligent control method |
| CN108584586A (en) * | 2017-11-21 | 2018-09-28 | 江苏鸿山鸿锦物联技术有限公司 | A kind of elevator door state line detection method based on video image |
| CN108584579A (en) * | 2018-04-24 | 2018-09-28 | 姜盎然 | A kind of intelligent elevator management control system and method based on passenger demand |
| CN110027958A (en) * | 2017-11-10 | 2019-07-19 | 奥的斯电梯公司 | It is determined for the model of safety and efficiency sensing and activity |
| CN110255318A (en) * | 2019-06-03 | 2019-09-20 | 浙江新再灵科技股份有限公司 | The method of the idle Articles detecting of lift car based on image, semantic segmentation |
| CN110510476A (en) * | 2018-05-22 | 2019-11-29 | 奥的斯电梯公司 | Pressure sensitive pads |
| CN110713082A (en) * | 2019-10-22 | 2020-01-21 | 日立楼宇技术（广州）有限公司 | Elevator control method, system, device and storage medium |
| CN110723614A (en) * | 2019-10-28 | 2020-01-24 | 珠海格力电器股份有限公司 | Protection device for elevator door, elevator and protection method |
| CN110927726A (en) * | 2019-11-14 | 2020-03-27 | 广东奥迪威传感科技股份有限公司 | Approach detection method and module |
| CN112850436A (en) * | 2019-11-28 | 2021-05-28 | 宁波微科光电股份有限公司 | Pedestrian trend detection method and system of elevator intelligent light curtain |
| CN113401774A (en) * | 2021-05-26 | 2021-09-17 | 杭州法维莱科技有限公司 | 90 degrees vertical hinged door systems of elevator with prevent pressing from both sides function |
| CN114348821A (en) * | 2021-12-06 | 2022-04-15 | 日立楼宇技术（广州）有限公司 | Method, system and device for guiding elevator passengers and storage medium |
| WO2024125589A1 (en) * | 2022-12-16 | 2024-06-20 | 日立楼宇技术（广州）有限公司 | Object behavior detection method and apparatus, electronic device and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000351560A (en) * | 1999-06-09 | 2000-12-19 | Mitsubishi Electric Building Techno Service Co Ltd | Door opening and closing control unit of elevator |
| JP2002293484A (en) * | 2001-03-29 | 2002-10-09 | Mitsubishi Electric Corp | Elevator control device |
| WO2010098024A1 (en) * | 2009-02-24 | 2010-09-02 | 三菱電機株式会社 | Human tracking device and human tracking program |
| CN101920896A (en) * | 2009-06-10 | 2010-12-22 | 株式会社日立制作所 | Lift facility |
| CN102036899A (en) * | 2008-05-22 | 2011-04-27 | 奥蒂斯电梯公司 | Video-based system and method of elevator door detection |
| CN202390065U (en) * | 2011-12-02 | 2012-08-22 | 广州都盛机电有限公司 | Elevator dynamic image recognition and control system |
-
2011
- 2011-12-12 CN CN2011104102354A patent/CN102701056A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000351560A (en) * | 1999-06-09 | 2000-12-19 | Mitsubishi Electric Building Techno Service Co Ltd | Door opening and closing control unit of elevator |
| JP2002293484A (en) * | 2001-03-29 | 2002-10-09 | Mitsubishi Electric Corp | Elevator control device |
| CN102036899A (en) * | 2008-05-22 | 2011-04-27 | 奥蒂斯电梯公司 | Video-based system and method of elevator door detection |
| WO2010098024A1 (en) * | 2009-02-24 | 2010-09-02 | 三菱電機株式会社 | Human tracking device and human tracking program |
| CN101920896A (en) * | 2009-06-10 | 2010-12-22 | 株式会社日立制作所 | Lift facility |
| CN202390065U (en) * | 2011-12-02 | 2012-08-22 | 广州都盛机电有限公司 | Elevator dynamic image recognition and control system |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110027958B (en) * | 2017-11-10 | 2020-11-06 | 奥的斯电梯公司 | Model sensing and activity determination for safety and efficiency |
| CN110027958A (en) * | 2017-11-10 | 2019-07-19 | 奥的斯电梯公司 | It is determined for the model of safety and efficiency sensing and activity |
| CN108584586A (en) * | 2017-11-21 | 2018-09-28 | 江苏鸿山鸿锦物联技术有限公司 | A kind of elevator door state line detection method based on video image |
| CN108529367A (en) * | 2017-11-21 | 2018-09-14 | 江苏鸿山鸿锦物联技术有限公司 | A kind of elevator door state patch detection method based on video image |
| CN108584579A (en) * | 2018-04-24 | 2018-09-28 | 姜盎然 | A kind of intelligent elevator management control system and method based on passenger demand |
| CN108529363A (en) * | 2018-04-26 | 2018-09-14 | 珠海亿智电子科技有限公司 | A kind of intelligent control method |
| CN110510476A (en) * | 2018-05-22 | 2019-11-29 | 奥的斯电梯公司 | Pressure sensitive pads |
| CN110255318A (en) * | 2019-06-03 | 2019-09-20 | 浙江新再灵科技股份有限公司 | The method of the idle Articles detecting of lift car based on image, semantic segmentation |
| CN110255318B (en) * | 2019-06-03 | 2021-03-09 | 浙江新再灵科技股份有限公司 | Method for detecting idle articles in elevator car based on image semantic segmentation |
| CN110713082A (en) * | 2019-10-22 | 2020-01-21 | 日立楼宇技术（广州）有限公司 | Elevator control method, system, device and storage medium |
| CN110723614A (en) * | 2019-10-28 | 2020-01-24 | 珠海格力电器股份有限公司 | Protection device for elevator door, elevator and protection method |
| CN110927726A (en) * | 2019-11-14 | 2020-03-27 | 广东奥迪威传感科技股份有限公司 | Approach detection method and module |
| CN112850436A (en) * | 2019-11-28 | 2021-05-28 | 宁波微科光电股份有限公司 | Pedestrian trend detection method and system of elevator intelligent light curtain |
| CN113401774A (en) * | 2021-05-26 | 2021-09-17 | 杭州法维莱科技有限公司 | 90 degrees vertical hinged door systems of elevator with prevent pressing from both sides function |
| CN114348821A (en) * | 2021-12-06 | 2022-04-15 | 日立楼宇技术（广州）有限公司 | Method, system and device for guiding elevator passengers and storage medium |
| CN114348821B (en) * | 2021-12-06 | 2023-09-12 | 日立楼宇技术（广州）有限公司 | Guiding method, system and device for elevator passengers and storage medium |
| WO2024125589A1 (en) * | 2022-12-16 | 2024-06-20 | 日立楼宇技术（广州）有限公司 | Object behavior detection method and apparatus, electronic device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102701056A (en) | Dynamic image recognition control method, device and system | |
| CN202390065U (en) | Elevator dynamic image recognition and control system | |
| CN101506077B (en) | Anonymous passenger indexing system for security tracking in destination entry dispatching operations | |
| CN102036899B (en) | Video-based system and method of elevator door detection | |
| US10189677B2 (en) | Elevator control system with facial recognition and authorized floor destination verification | |
| JP6657167B2 (en) | User detection system | |
| US10706553B2 (en) | Image detection device for detecting shadow or light based on detected edges | |
| CN109063764A (en) | A kind of judgment method of disconnecting switch closing operation in place based on machine vision | |
| CN104794458A (en) | Fuzzy video person identifying method | |
| CN101105097A (en) | Platform shield door intelligent control method | |
| CN110032966A (en) | Human body proximity test method, intelligent Service method and device for intelligent Service | |
| CN107902513A (en) | A kind of elevator based on video/image detection stops control method | |
| CN110969131B (en) | Subway people flow counting method based on scene flow | |
| JP6139729B1 (en) | Image processing device | |
| CN111539259A (en) | Target object recognition method, artificial neural network training method, computer-readable storage medium, and manipulator | |
| KR101066562B1 (en) | Screen door system using stereo image | |
| KR20170052285A (en) | Intelligent camera and System for automatic opening door using Intelligent camera | |
| CN108178031B (en) | Stretcher mode identification method, device and system in lift car | |
| Han et al. | Adaptive background modeling with shadow suppression | |
| CN111104554B (en) | Automatic door control method and system based on contour detection and head and shoulder detection | |
| Patel et al. | Human tracking in video surveillance | |
| CN205334579U (en) | Intelligence garage parking system | |
| CN103366598B (en) | A kind of vehicle parking is to position detecting system and detection method | |
| JP7248203B2 (en) | Elevator device and falling object detection device | |
| CN115423764A (en) | Method and device for detecting shielding of camera and computer equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20121003 |