CN105447811A - Spatial data processing method, device and system - Google Patents
Spatial data processing method, device and system Download PDFInfo
- Publication number
- CN105447811A CN105447811A CN201510819538.XA CN201510819538A CN105447811A CN 105447811 A CN105447811 A CN 105447811A CN 201510819538 A CN201510819538 A CN 201510819538A CN 105447811 A CN105447811 A CN 105447811A
- Authority
- CN
- China
- Prior art keywords
- algorithm
- equipment
- spatial data
- gpu
- cpu
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T1/00—General purpose image data processing
- G06T1/20—Processor architectures; Processor configuration, e.g. pipelining
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2200/00—Indexing scheme for image data processing or generation, in general
- G06T2200/28—Indexing scheme for image data processing or generation, in general involving image processing hardware
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a spatial data processing method, and the method comprises the steps: obtaining to-be-processed spatial data, and determining an algorithm according to the processing type of the to-be-processed spatial data; selecting a CPU or GPU for the determined algorithm, and controlling the CPU and GPU to carry out the cooperative processing of the to-be-processed spatial data according to the determined algorithm; and receiving the processing result returned by the CPU/GPU, so as to obtain the final processing result of the to-be-processed spatial data. The invention also discloses a spatial data processing device and system. According to the invention, the method, device and system can improve the processing efficiency of spatial data and reduce the cost.
Description
Technical field
The present invention relates to data processing field, particularly relate to a kind of spatial data processing method, Apparatus and system.
Background technology
Spatial data refers to the data for the position of representation space entity, shape, size and all multi-aspect informations of distribution characteristics thereof, and it can be used for describing the target from real world, and it has the characteristics such as location, qualitative, Time and place relation.Spatial data is a kind of data representing the natural world that people depend on for existence by the fundamental space such as point, line, surface and entity data structure.Common spatial data comprises: vector data (point, line, surface, body) and raster data (image, elevation).
The luv space data that collection obtains cannot directly use, and they need, under many process can reach reasonable effect and consistent coordinate system is arrived in unification, to only have the spatial data after processing like this can be used for various embody rule.
Along with the progress of technology, the means of space data collection are also updated, full-automatic remote sensing technology is surveyed and drawn from artificial pointwise, the speed goes that space data collection obtains is fast, meanwhile all trades and professions also grow with each passing day for the demand of spatial data, the increase of spatial data demand is embodied in two aspects, is first the sharply increase of spatial data coverage and data volume, is secondly the sharply increase of spatial data requirement of real-time.
Change in the demand of spatial data proposes high requirement for the speed of spatial data handling, in the prior art, all adopt central processor CPU equipment to process spatial data, because the optimization of algorithm reaches bottleneck, therefore the processing speed of spatial data does not more and more catch up with the change of demand.For above-mentioned defect, settling mode of the prior art is: adopt the mode of multiple stage CPU equipment collaboration calculating to process spatial data, polylith is divided into by spatial data, respectively by tens or hundreds of platform CPU device processes, then the result of each CPU equipment is merged and obtain complete result, although this mode can improve the treatment effeciency of spatial data to a certain extent, but its technical threshold and high cost, due to adopt tens or hundreds of platform CPU equipment carry out spatial data handling, high for the requirement of building whole computing environment, and it is also high to build cost.
Foregoing, only for auxiliary understanding technical scheme of the present invention, does not represent and admits that foregoing is prior art.
Summary of the invention
Fundamental purpose of the present invention is to provide a kind of spatial data processing method, Apparatus and system, is intended to solve in prior art, when processing spatial data, and the lower and technical matters that cost is high for the treatment of effeciency.
For achieving the above object, the invention provides a kind of spatial data processing method, the method comprises:
Obtain pending spatial data, and according to the process type determination algorithm of described pending spatial data;
For the described algorithms selection central processor CPU equipment determined or graphic process unit GPU equipment, and control described CPU equipment and GPU equipment carries out associated treatment according to the algorithm determined to pending spatial data;
Receive the result that described CPU equipment and/or GPU equipment return, to obtain the final process result to described pending spatial data.
Preferably, described is the described algorithms selection central processor CPU equipment determined or graphic process unit GPU equipment, and controls described CPU equipment and GPU equipment comprises the step that pending spatial data carries out associated treatment according to the algorithm determined:
According to the process exponential sum process index in described GPU equipment of the described algorithm determined in described CPU equipment, select CPU equipment that the described algorithm determined is corresponding or GPU equipment, and control described CPU equipment and GPU equipment carries out associated treatment according to the algorithm determined to pending spatial data.
Preferably, described is the described algorithms selection central processor CPU equipment determined or graphic process unit GPU equipment, and controls described CPU equipment and GPU equipment comprises the step that pending spatial data carries out associated treatment according to the algorithm determined:
According to the algorithm preset and the mapping relations of actuating equipment, find CPU equipment corresponding to the described algorithm determined or GPU equipment, and control described CPU equipment and GPU equipment carries out associated treatment according to the algorithm determined to pending spatial data.
Preferably, described pending spatial data is pending remotely-sensed data; The described algorithm determined comprises the even smooth algorithm of coordinate transformation algorithm, projective transformation algorithm, image resampling algorithm, scalloping algorithm, Image Fusion, extraction characteristic curve algorithm and image.
Preferably, describedly for the described algorithms selection central processor CPU equipment determined or graphic process unit GPU equipment be: be described coordinate transformation algorithm, projective transformation algorithm, image resampling algorithm and scalloping algorithms selection GPU equipment, be described Image Fusion, extract characteristic curve algorithm and the even smooth algorithms selection CPU equipment of image.
In addition, for achieving the above object, the present invention also provides a kind of spatial data handling device, and this device comprises:
Acquisition module, for obtaining pending spatial data, according to the process type determination algorithm of described pending spatial data;
Select control module, for being the described algorithms selection central processor CPU equipment determined or graphic process unit GPU equipment, and control described CPU equipment and GPU equipment carries out associated treatment according to the algorithm determined to pending spatial data;
Receiver module, for receiving the result that described CPU equipment and/or GPU equipment return, to obtain the final process result to described pending spatial data.
Preferably, described selection control module is also for according to the process exponential sum process index in described GPU equipment of the described algorithm determined at described CPU equipment, select CPU equipment that the described algorithm determined is corresponding or GPU equipment, and control described CPU equipment and GPU equipment carries out associated treatment according to the algorithm determined to pending spatial data.
Preferably, described selection control module is also for the mapping relations according to the algorithm preset and actuating equipment, find CPU equipment corresponding to the described algorithm determined or GPU equipment, and control described CPU equipment and GPU equipment carries out associated treatment according to the algorithm determined to pending spatial data.
Preferably, described pending spatial data is pending remotely-sensed data; The described algorithm determined comprises the even smooth algorithm of coordinate transformation algorithm, projective transformation algorithm, image resampling algorithm, scalloping algorithm, Image Fusion, extraction characteristic curve algorithm and image.
Preferably, described selection control module, also for being described coordinate transformation algorithm, projective transformation algorithm, image resampling algorithm and scalloping algorithms selection GPU equipment, for described Image Fusion, extracts characteristic curve algorithm and the even smooth algorithms selection CPU equipment of image.
In addition, for achieving the above object, the present invention also provides a kind of spatial data handling system, comprises above-mentioned spatial data handling device, and the central processor CPU equipment, the graphic process unit GPU equipment that are connected respectively with described spatial data handling device.
Spatial data processing method of the present invention, Apparatus and system, by obtaining pending spatial data, and according to the process type determination algorithm of described pending spatial data; For the described algorithms selection central processor CPU equipment determined or graphic process unit GPU equipment, and control described CPU equipment and GPU equipment carries out associated treatment according to the algorithm determined to pending spatial data; Receive the result that described CPU equipment and/or GPU equipment return, to obtain the final process result to described pending spatial data; Namely be first this pending spatial data determination algorithm, then the algorithm for determining determines corresponding actuating equipment (CPU equipment or GPU equipment), then by the actuating equipment determined, associated treatment is carried out to pending spatial data, the result returned according to actuating equipment again obtains the final process result of pending spatial data, the treatment effeciency of pending spatial data can be improved, reduce the stand-by period, also can reduce costs.
Accompanying drawing explanation
Fig. 1 is the schematic flow sheet of the preferred embodiment of spatial data processing method of the present invention;
Fig. 2 is the structural representation of the preferred embodiment of spatial data handling device of the present invention;
Fig. 3 is the structural representation of the preferred embodiment of spatial data handling system of the present invention.
The realization of the object of the invention, functional characteristics and advantage will in conjunction with the embodiments, are described further with reference to accompanying drawing.
Embodiment
Should be appreciated that specific embodiment described herein only in order to explain the present invention, be not intended to limit the present invention.
With reference to the schematic flow sheet that Fig. 1, Fig. 1 are the preferred embodiment of spatial data processing method of the present invention, the method comprises:
S10, obtain pending spatial data, and according to the process type determination algorithm of this pending spatial data.
In this step, can obtain pending spatial data from spatial data acquisition device, this pending spatial data can be vector data (point, line, surface, body) and raster data (image, elevation).
In this step, according to the process type determination algorithm of this pending spatial data, the algorithm that the pending spatial data of different disposal type is corresponding is different, the algorithm that the process type of pending spatial data is corresponding can be pre-set, have algorithm 1 and algorithm 2 as processed algorithm corresponding to type A, the algorithm processing type B corresponding has algorithm 3, algorithm 4 and algorithm 5.This process type comprises: the collision detection of the choosing of spatial data, abbreviation, transmission and element annotation with dodge.The process type of this pending spatial data can be one or more.The process type of this pending spatial data can preset.In one embodiment, this algorithm determined comprises coordinates transformation method, projective transformation algorithm, and if algorithm 1 is coordinates transformation method, algorithm 2 is projection change algorithm.
The algorithm that the process type of this pending spatial data is corresponding has one at least, be generally multiple, as comprised algorithm 1, algorithm 2, algorithm 3, algorithm 4 and algorithm 5, between each algorithm, some can parallel processing, some can only serial processing, and as algorithm 1 and algorithm 2 can process simultaneously, algorithm 3 is in algorithm 1 aftertreatment (as algorithm 3 is relevant to algorithm 4), algorithm 4 is in algorithm 2 aftertreatment (as algorithm 4 is relevant to algorithm 2), and algorithm 5 is in algorithm 3 and algorithm 4 aftertreatment.Can the algorithm of parallel processing can be performed by different actuating equipments (CPU equipment or GPU equipment), to improve treatment effeciency respectively simultaneously.
S20, be this algorithms selection central processor CPU equipment determined or graphic process unit GPU equipment, and control this CPU equipment and GPU equipment carries out associated treatment according to the algorithm determined to pending spatial data.
For the algorithms selection actuating equipment that this is determined, this algorithm determined can select one as actuating equipment from CPU equipment and GPU equipment, common, for this algorithm determined selects an execution efficiency high from CPU equipment and GPU equipment, such as, the algorithm determined is algorithm 1, algorithm 1 all can perform in CPU equipment and GPU equipment, but the execution efficiency in GPU equipment is higher, then can be this algorithm 1 and select GPU equipment, by this GPU equipment execution algorithm 1.
In this step, for the algorithms selection determined selects central processor CPU equipment or graphic process unit GPU equipment, as in one embodiment, the algorithm determined comprises algorithm 1, algorithm 2, algorithm 3, algorithm 4 and algorithm 5, for algorithm 1 selects CPU equipment, for algorithm 2 selects GPU equipment, for algorithm 3 selects CPU equipment, for algorithm 4 selects GPU equipment, for algorithm 5 selects GPU equipment, be CPU equipment and GPU equipment is assigned with the algorithm process needing to carry out pending spatial data.In this step, also control this this CPU equipment and GPU equipment carries out associated treatment according to the algorithm determined to pending spatial data, as in one embodiment, control this CPU equipment to carry out process according to algorithm 1 to pending spatial data and obtain the first result, control this GPU equipment simultaneously and according to algorithm 2, process is carried out to pending spatial data and obtain the second result; Control CPU equipment carries out process according to algorithm 3 to the first result and obtains the 3rd result again, and control GPU equipment carries out process according to algorithm 4 to the second result and obtains the 4th result simultaneously; Control GPU equipment carries out process according to algorithm 5 to the 3rd result and the 4th result and obtains the 5th result again; Make this CPU equipment and GPU equipment carry out associated treatment to pending spatial data, improve the treatment effeciency of pending spatial data, reduce the stand-by period.
S30, receive the result that this CPU equipment and/or GPU equipment returns, to obtain the final process result to this pending spatial data.
In this step, from CPU equipment and/or GPU equipment, result is obtained, namely to the final process result of pending spatial data.When last algorithm of this CPU equipment and GPU equipment is in concurrent operation (last algorithm of CPU equipment and GPU equipment is separate), then receive the result that this CPU equipment and GPU equipment return, and the result that the result returned of this CPU equipment and this GPU equipment return is merged, obtain the final process result to this pending spatial data, when last algorithm of this CPU equipment and GPU equipment is not when being in parallel running, then determine to receive the result returned from this CPU equipment and GPU equipment according to the priority execution sequence between this CPU equipment and last algorithm of GPU equipment, as in the above-described embodiments, the algorithm determined has 5, last algorithm that GPU equipment performs is algorithm 5, last algorithm that CPU performs is algorithm 3, this algorithm 5 performs after algorithm 4, then determine to receive the result returned from this GPU equipment, obtain the final process result to this pending spatial data.
Adopt above-described embodiment, by obtaining pending spatial data, and according to the process type determination algorithm of this pending spatial data; The algorithms selection central processor CPU equipment determined for this or graphic process unit GPU equipment, and control this CPU equipment and GPU equipment carries out associated treatment according to the algorithm determined to pending spatial data; Receive the result that this CPU equipment and/or GPU equipment return, to obtain the final process result to this pending spatial data; Namely be first this pending spatial data determination algorithm, then the algorithm for determining determines corresponding actuating equipment (CPU equipment or GPU equipment), then by the actuating equipment determined, associated treatment is carried out to pending spatial data, the result returned according to actuating equipment again obtains the final process result of pending spatial data, the treatment effeciency of pending spatial data can be improved, reduce the stand-by period, also can reduce costs.
Further, should be this algorithms selection central processor CPU equipment determined or graphic process unit GPU equipment, and control this CPU equipment and GPU equipment comprises the step that pending spatial data carries out associated treatment according to the algorithm determined:
According to the process exponential sum process index in this GPU equipment of this algorithm determined in this CPU equipment, the CPU equipment that the algorithm selecting this to determine is corresponding or GPU equipment, and control this CPU equipment and GPU equipment carries out associated treatment according to the algorithm determined to pending spatial data.
Can test in CPU equipment and GPU equipment the algorithm determined in advance, obtain the process index in the algorithm the determined process exponential sum GPU equipment in CPU equipment, algorithm as determined is algorithm 1, the process index of algorithm 1 in CPU equipment is 1, the process index of algorithm 1 in GPU equipment is 2, process index is larger, shows that treatment effeciency is higher.
In this step, at the process exponential sum process index in this GPU equipment of the algorithm determined according to this in this CPU equipment, when the CPU equipment that the algorithm selecting this to determine is corresponding or GPU equipment, select the high CPU equipment of process index or GPU equipment as actuating equipment corresponding to this algorithm determined, if the process index of algorithm 1 in CPU equipment is higher than the process index in GPU equipment, then select CPU equipment as the actuating equipment of correspondence for this algorithm 1, if the process index of algorithm 2 in GPU equipment is higher than the process index in CPU equipment, then select GPU equipment as the actuating equipment of correspondence for algorithm 2.
Further, should be this algorithms selection central processor CPU equipment determined or graphic process unit GPU equipment, and control this CPU equipment and GPU equipment comprises the step that pending spatial data carries out associated treatment according to the algorithm determined:
According to the algorithm preset and the mapping relations of actuating equipment, find CPU equipment corresponding to this algorithm determined or GPU equipment, and control this CPU equipment and GPU equipment carries out associated treatment according to the algorithm determined to pending spatial data.
The mapping relations of algorithm and actuating equipment can be preset, if the actuating equipment arranging algorithm 1 correspondence is CPU equipment, the actuating equipment of algorithm 2 correspondence is GPU equipment, the actuating equipment of algorithm 3 correspondence is CPU equipment, the actuating equipment of algorithm 4 correspondence is GPU equipment, the actuating equipment of algorithm 5 correspondence is CPU equipment, etc.Supvr can upgrade the mapping relations of this algorithm and actuating equipment in real time.
In this step, according to the algorithm that this is determined, travel through the mapping relations of this algorithm and actuating equipment, find CPU equipment corresponding to this algorithm determined or GPU equipment, when algorithm as determined is algorithm 3, then finding actuating equipment corresponding to this algorithm determined is CPU equipment; When algorithm as determined is algorithm 4, then finding actuating equipment corresponding to this algorithm determined is GPU equipment.Adopting aforesaid way, can be conveniently actuating equipment that the algorithms selection determined is corresponding.
Further, this pending spatial data is pending remotely-sensed data; This algorithm determined comprises the even smooth algorithm of coordinate transformation algorithm, projective transformation algorithm, image resampling algorithm, scalloping algorithm, Image Fusion, extraction characteristic curve algorithm and image.
In one embodiment, should be this algorithms selection central processor CPU equipment determined or graphic process unit GPU equipment, and control this CPU equipment and GPU equipment according to the algorithm determined to the step that pending spatial data carries out associated treatment is: be this coordinate transformation algorithm, projective transformation algorithm, image resampling algorithm and scalloping algorithms selection GPU equipment, be this Image Fusion, extract characteristic curve algorithm and the even smooth algorithms selection CPU equipment of image; The execution sequence of the algorithm (the even smooth algorithm of coordinate transformation algorithm, projective transformation algorithm, image resampling algorithm, scalloping algorithm, Image Fusion, extraction characteristic curve algorithm and image) determined is followed successively by: coordinate transformation algorithm, projective transformation algorithm, image resampling algorithm, extraction characteristic curve algorithm, scalloping algorithm, Image Fusion, the even smooth algorithm of image; In GPU equipment, input needs vector data or the raster data of conversion, adopts coordinate transformation algorithm and the data of projective transformation algorithm to input to carry out coordinate space transformations, and the vector data exported through coordinate space transformations or raster data; In GPU equipment, input pending raster data (raster data through coordinate space transformations), adopt image resampling algorithm to carry out resampling process to pending raster data, generate the first raster data after resampling; In CPU equipment, input the first raster data (this first raster data has several), adopt extraction characteristic curve algorithm to process this first raster data, identify same area on several raster datas, the characteristic curve of output identification same area; Input in GPU equipment pending raster data (through extraction characteristic curve algorithm the first raster data), adopt scalloping algorithm to carry out deformation process to pending raster data, generate the second raster data after distortion; In CPU equipment, input pending raster data (this raster data has several), the second raster data described above, adopt Image Fusion to be merged by pending each width raster data, export the 3rd raster data after merging; In CPU equipment, the image data (the 3rd raster data described above) of even light is treated in input, adopts in the even smooth algorithm process image data of image and exposes unbalanced region, exports the image data of exposure color balancing.
The result that this CPU equipment of this reception and/or GPU equipment return, obtaining to the step of the final process result of this pending spatial data be: receive the result that this CPU equipment returns, obtain the final process result to this pending remotely-sensed data.
With reference to the structural representation that Fig. 2, Fig. 2 are the preferred embodiment of spatial data handling device of the present invention, this device 100 comprises:
Acquisition module 10, for obtaining pending spatial data, according to the process type determination algorithm of this pending spatial data;
Select control module 20, for being this algorithms selection central processor CPU equipment determined or graphic process unit GPU equipment, and control this CPU equipment and GPU equipment carries out associated treatment according to the algorithm determined to pending spatial data;
Receiver module 30, for receiving the result that this CPU equipment and/or GPU equipment return.
This acquisition module 10 can obtain pending spatial data from spatial data acquisition device, and this pending spatial data can be vector data (point, line, surface, body) and raster data (image, elevation).
This acquisition module 10 is according to the process type determination algorithm of this pending spatial data, the algorithm that the pending spatial data of different disposal type is corresponding is different, the algorithm that the process type of pending spatial data is corresponding can be pre-set, have algorithm 1 and algorithm 2 as processed algorithm corresponding to type A, the algorithm processing type B corresponding has algorithm 3, algorithm 4 and algorithm 5.This process type comprises: the collision detection of the choosing of spatial data, abbreviation, transmission and element annotation with dodge.The process type of this pending spatial data can be one or more.The process type of this pending spatial data can preset.In one embodiment, this algorithm determined comprises coordinates transformation method, projective transformation algorithm, and if algorithm 1 is coordinates transformation method, algorithm 2 is projection change algorithm.
The algorithm that the process type of this pending spatial data is corresponding has one at least, be generally multiple, as comprised algorithm 1, algorithm 2, algorithm 3, algorithm 4 and algorithm 5, between each algorithm, some can parallel processing, some can only serial processing, and as algorithm 1 and algorithm 2 can process simultaneously, algorithm 3 is in algorithm 1 aftertreatment (as algorithm 3 is relevant to algorithm 4), algorithm 4 is in algorithm 2 aftertreatment (as algorithm 4 is relevant to algorithm 2), and algorithm 5 is in algorithm 3 and algorithm 4 aftertreatment.Can the algorithm of parallel processing can be performed by different actuating equipments (CPU equipment or GPU equipment), to improve treatment effeciency respectively simultaneously.
The algorithms selection actuating equipment that this selection control module 20 is determined for this, this algorithm determined can select one as actuating equipment from CPU equipment and GPU equipment, common, for this algorithm determined selects an execution efficiency high from CPU equipment and GPU equipment, such as, the algorithm determined is algorithm 1, algorithm 1 all can perform in CPU equipment and GPU equipment, but the execution efficiency in GPU equipment is higher, then can be this algorithm 1 and select GPU equipment, by this GPU equipment execution algorithm 1.
This selection control module 20 selects central processor CPU equipment or graphic process unit GPU equipment for the algorithms selection determined, as in one embodiment, the algorithm determined comprises algorithm 1, algorithm 2, algorithm 3, algorithm 4 and algorithm 5, for algorithm 1 selects CPU equipment, for algorithm 2 selects GPU equipment, for algorithm 3 selects CPU equipment, for algorithm 4 selects GPU equipment, for algorithm 5 selects GPU equipment, be CPU equipment and GPU equipment is assigned with the algorithm process needing to carry out pending spatial data.In this step, also control this this CPU equipment and GPU equipment carries out associated treatment according to the algorithm determined to pending spatial data, as in one embodiment, control this CPU equipment to carry out process according to algorithm 1 to pending spatial data and obtain the first result, control this GPU equipment simultaneously and according to algorithm 2, process is carried out to pending spatial data and obtain the second result; Control CPU equipment carries out process according to algorithm 3 to the first result and obtains the 3rd result again, and control GPU equipment carries out process according to algorithm 4 to the second result and obtains the 4th result simultaneously; Control GPU equipment carries out process according to algorithm 5 to the 3rd result and the 4th result and obtains the 5th result again; Make this CPU equipment and GPU equipment carry out associated treatment to pending spatial data, improve the treatment effeciency of pending spatial data, reduce the stand-by period.
This receiver module 30 obtains result from CPU equipment and/or GPU equipment, namely to the final process result of pending spatial data.When last algorithm of this CPU equipment and GPU equipment is in concurrent operation (last algorithm of CPU equipment and GPU equipment is separate), then this receiver module 30 receives the result that this CPU equipment and GPU equipment return, and the result that the result returned of this CPU equipment and this GPU equipment return is merged, obtain the final process result to this pending spatial data, when last algorithm of this CPU equipment and GPU equipment is not when being in parallel running, then this receiver module 30 is determined to receive the result returned from this CPU equipment and GPU equipment according to the priority execution sequence between this CPU equipment and last algorithm of GPU equipment, as in the above-described embodiments, the algorithm determined has 5, last algorithm that GPU equipment performs is algorithm 5, last algorithm that CPU performs is algorithm 3, this algorithm 5 performs after algorithm 4, then determine to receive the result returned from this GPU equipment, obtain the final process result to this pending spatial data.
Further, this selection control module 20 is also for the algorithm determined according to this process index of process exponential sum in this GPU equipment at this CPU equipment, the CPU equipment that the algorithm selecting this to determine is corresponding or GPU equipment, and control this CPU equipment and GPU equipment carries out associated treatment according to the algorithm determined to pending spatial data.
Can test in CPU equipment and GPU equipment the algorithm determined in advance, obtain the process index in the algorithm the determined process exponential sum GPU equipment in CPU equipment, algorithm as determined is algorithm 1, the process index of algorithm 1 in CPU equipment is 1, the process index of algorithm 1 in GPU equipment is 2, process index is larger, shows that treatment effeciency is higher.
This selection control module 20 is at the process exponential sum process index in this GPU equipment of the algorithm determined according to this in this CPU equipment, when the CPU equipment that the algorithm selecting this to determine is corresponding or GPU equipment, select the high CPU equipment of process index or GPU equipment as actuating equipment corresponding to this algorithm determined, if the process index of algorithm 1 in CPU equipment is higher than the process index in GPU equipment, then select CPU equipment as the actuating equipment of correspondence for this algorithm 1, if the process index of algorithm 2 in GPU equipment is higher than the process index in CPU equipment, then select GPU equipment as the actuating equipment of correspondence for algorithm 2.
Further, this selection control module 20 is also for the mapping relations according to the algorithm preset and actuating equipment, find CPU equipment corresponding to this algorithm determined or GPU equipment, and control this CPU equipment and GPU equipment carries out associated treatment according to the algorithm determined to pending spatial data.
According to the algorithm preset and the mapping relations of actuating equipment, find CPU equipment corresponding to this algorithm determined or GPU equipment, and control this CPU equipment and GPU equipment carries out associated treatment according to the algorithm determined to pending spatial data.
The mapping relations of algorithm and actuating equipment can be preset, if the actuating equipment arranging algorithm 1 correspondence is CPU equipment, the actuating equipment of algorithm 2 correspondence is GPU equipment, the actuating equipment of algorithm 3 correspondence is CPU equipment, the actuating equipment of algorithm 4 correspondence is GPU equipment, the actuating equipment of algorithm 5 correspondence is CPU equipment, etc.Supvr can upgrade the mapping relations of this algorithm and actuating equipment in real time.
The algorithm that this selection control module 20 is determined according to this, travel through the mapping relations of this algorithm and actuating equipment, find CPU equipment corresponding to this algorithm determined or GPU equipment, when the algorithm as determined is algorithm 3, then finding actuating equipment corresponding to this algorithm determined is CPU equipment; When algorithm as determined is algorithm 4, then finding actuating equipment corresponding to this algorithm determined is GPU equipment.Adopting aforesaid way, can be conveniently actuating equipment that the algorithms selection determined is corresponding.
Further, this pending spatial data is pending remotely-sensed data; This algorithm determined comprises the even smooth algorithm of coordinate transformation algorithm, projective transformation algorithm, image resampling algorithm, scalloping algorithm, Image Fusion, extraction characteristic curve algorithm and image.
In one embodiment, this selection control module 20 is this coordinate transformation algorithm, projective transformation algorithm, image resampling algorithm and scalloping algorithms selection GPU equipment, for this Image Fusion, extracts characteristic curve algorithm and the even smooth algorithms selection CPU equipment of image; The execution sequence of the algorithm (the even smooth algorithm of coordinate transformation algorithm, projective transformation algorithm, image resampling algorithm, scalloping algorithm, Image Fusion, extraction characteristic curve algorithm and image) determined is followed successively by: coordinate transformation algorithm, projective transformation algorithm, image resampling algorithm, extraction characteristic curve algorithm, scalloping algorithm, Image Fusion, the even smooth algorithm of image; In GPU equipment, input needs vector data or the raster data of conversion, adopts coordinate transformation algorithm and the data of projective transformation algorithm to input to carry out coordinate space transformations, and the vector data exported through coordinate space transformations or raster data; In GPU equipment, input pending raster data (raster data through coordinate space transformations), adopt image resampling algorithm to carry out resampling process to pending raster data, generate the first raster data after resampling; In CPU equipment, input the first raster data (this first raster data has several), adopt extraction characteristic curve algorithm to process this first raster data, identify same area on several raster datas, the characteristic curve of output identification same area; Input in GPU equipment pending raster data (through extraction characteristic curve algorithm the first raster data), adopt scalloping algorithm to carry out deformation process to pending raster data, generate the second raster data after distortion; In CPU equipment, input pending raster data (this raster data has several), the second raster data described above, adopt Image Fusion to be merged by pending each width raster data, export the 3rd raster data after merging; In CPU equipment, the image data (the 3rd raster data described above) of even light is treated in input, adopts in the even smooth algorithm process image data of image and exposes unbalanced region, exports the image data of exposure color balancing.
This receiver module 30 receives the result that this CPU equipment returns, and obtains the final process result to this pending remotely-sensed data.
As shown in Figure 3, Fig. 3 is the structural representation of the preferred embodiment of spatial data handling system of the present invention, this system comprises the spatial data handling device 100 of above-described embodiment, and the central processor CPU equipment 200, the graphic process unit GPU equipment 300 that are connected respectively with this spatial data handling device 100.
This spatial data handling device 100 for obtaining pending spatial data, and according to the process type determination algorithm of this pending spatial data; And for being this algorithms selection central processor CPU equipment 200 determined or graphic process unit GPU equipment 300, and control this CPU equipment 200 and GPU equipment 300 carries out associated treatment according to the algorithm determined to pending spatial data; And for receiving the result that this CPU equipment 200 and/or GPU equipment 300 return, to obtain the final process result to this pending spatial data;
This CPU equipment 200 and GPU equipment 300 carry out associated treatment according to the control of spatial data handling device 100 to pending spatial data, improve treatment effeciency.
These are only the preferred embodiments of the present invention; not thereby the scope of the claims of the present invention is limited; every utilize instructions of the present invention and accompanying drawing content to do equivalent structure or equivalent flow process conversion; or be directly or indirectly used in other relevant technical fields, be all in like manner included in scope of patent protection of the present invention.
Claims (10)
1. a spatial data processing method, is characterized in that, the method comprises:
Obtain pending spatial data, and according to the process type determination algorithm of described pending spatial data;
For the described algorithms selection central processor CPU equipment determined or graphic process unit GPU equipment, and control described CPU equipment and GPU equipment carries out associated treatment according to the algorithm determined to pending spatial data;
Receive the result that described CPU equipment and/or GPU equipment return, to obtain the final process result to described pending spatial data.
2. spatial data processing method as claimed in claim 1, it is characterized in that, described is the described algorithms selection central processor CPU equipment determined or graphic process unit GPU equipment, and controls described CPU equipment and GPU equipment comprises the step that pending spatial data carries out associated treatment according to the algorithm determined:
According to the process exponential sum process index in described GPU equipment of the described algorithm determined in described CPU equipment, select CPU equipment that the described algorithm determined is corresponding or GPU equipment, and control described CPU equipment and GPU equipment carries out associated treatment according to the algorithm determined to pending spatial data.
3. spatial data processing method as claimed in claim 1, it is characterized in that, described is the described algorithms selection central processor CPU equipment determined or graphic process unit GPU equipment, and controls described CPU equipment and GPU equipment comprises the step that pending spatial data carries out associated treatment according to the algorithm determined:
According to the algorithm preset and the mapping relations of actuating equipment, find CPU equipment corresponding to the described algorithm determined or GPU equipment, and control described CPU equipment and GPU equipment carries out associated treatment according to the algorithm determined to pending spatial data.
4. the spatial data processing method as described in any one of claims 1 to 3, is characterized in that, described pending spatial data is pending remotely-sensed data; The described algorithm determined comprises the even smooth algorithm of coordinate transformation algorithm, projective transformation algorithm, image resampling algorithm, scalloping algorithm, Image Fusion, extraction characteristic curve algorithm and image.
5. spatial data processing method as claimed in claim 4, it is characterized in that, describedly for the described algorithms selection central processor CPU equipment determined or graphic process unit GPU equipment be: be described coordinate transformation algorithm, projective transformation algorithm, image resampling algorithm and scalloping algorithms selection GPU equipment, be described Image Fusion, extract characteristic curve algorithm and the even smooth algorithms selection CPU equipment of image.
6. a spatial data handling device, is characterized in that, this system comprises:
Acquisition module, for obtaining pending spatial data, according to the process type determination algorithm of described pending spatial data;
Select control module, for being the described algorithms selection central processor CPU equipment determined or graphic process unit GPU equipment, and control described CPU equipment and GPU equipment carries out associated treatment according to the algorithm determined to pending spatial data;
Receiver module, for receiving the result that described CPU equipment and/or GPU equipment return, to obtain the final process result to described pending spatial data.
7. spatial data handling device as claimed in claim 6, it is characterized in that, described selection control module is also for according to the process exponential sum process index in described GPU equipment of the described algorithm determined at described CPU equipment, select CPU equipment that the described algorithm determined is corresponding or GPU equipment, and control described CPU equipment and GPU equipment carries out associated treatment according to the algorithm determined to pending spatial data.
8. spatial data handling device as claimed in claim 6, it is characterized in that, described selection control module is also for the mapping relations according to the algorithm preset and actuating equipment, find CPU equipment corresponding to the described algorithm determined or GPU equipment, and control described CPU equipment and GPU equipment carries out associated treatment according to the algorithm determined to pending spatial data.
9. spatial data handling device as claimed in claim 8, it is characterized in that, described pending spatial data is pending remotely-sensed data; The described algorithm determined comprises coordinate transformation algorithm, projective transformation algorithm, image resampling algorithm, scalloping algorithm, Image Fusion and extraction characteristic curve algorithm, the even smooth algorithm of image.
10. a spatial data handling system, is characterized in that, comprises the spatial data handling device described in any one of claim 6-9, and the central processor CPU equipment, the graphic process unit GPU equipment that are connected respectively with described spatial data handling device.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510819538.XA CN105447811A (en) | 2015-11-23 | 2015-11-23 | Spatial data processing method, device and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510819538.XA CN105447811A (en) | 2015-11-23 | 2015-11-23 | Spatial data processing method, device and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105447811A true CN105447811A (en) | 2016-03-30 |
Family
ID=55557945
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510819538.XA Pending CN105447811A (en) | 2015-11-23 | 2015-11-23 | Spatial data processing method, device and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105447811A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017210822A1 (en) * | 2016-06-06 | 2017-12-14 | Sz Dji Osmo Technology Co., Ltd. | Image processing for tracking |
| CN112256195A (en) * | 2020-10-19 | 2021-01-22 | 安徽工业大学 | Financial data storage method and system based on GPU |
| US11106928B2 (en) | 2016-06-06 | 2021-08-31 | Sz Dji Osmo Technology Co., Ltd. | Carrier-assisted tracking |
| CN115297359A (en) * | 2022-07-29 | 2022-11-04 | 北京字跳网络技术有限公司 | Multimedia data transmission method and device, electronic equipment and storage medium |
-
2015
- 2015-11-23 CN CN201510819538.XA patent/CN105447811A/en active Pending
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017210822A1 (en) * | 2016-06-06 | 2017-12-14 | Sz Dji Osmo Technology Co., Ltd. | Image processing for tracking |
| US10902609B2 (en) | 2016-06-06 | 2021-01-26 | Sz Dji Osmo Technology Co., Ltd. | Image processing for tracking |
| US11106928B2 (en) | 2016-06-06 | 2021-08-31 | Sz Dji Osmo Technology Co., Ltd. | Carrier-assisted tracking |
| US11568626B2 (en) | 2016-06-06 | 2023-01-31 | Sz Dji Osmo Technology Co., Ltd. | Carrier-assisted tracking |
| CN112256195A (en) * | 2020-10-19 | 2021-01-22 | 安徽工业大学 | Financial data storage method and system based on GPU |
| CN115297359A (en) * | 2022-07-29 | 2022-11-04 | 北京字跳网络技术有限公司 | Multimedia data transmission method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11422261B2 (en) | Robot relocalization method and apparatus and robot using the same | |
| CN105447811A (en) | Spatial data processing method, device and system | |
| US11111785B2 (en) | Method and device for acquiring three-dimensional coordinates of ore based on mining process | |
| CN112884764B (en) | Method and device for extracting land block in image, electronic equipment and storage medium | |
| CN105069423A (en) | Human body posture detection method and device | |
| CN111582123B (en) | AGV positioning method based on beacon identification and visual SLAM | |
| CN108197567B (en) | Method, apparatus and computer readable medium for image processing | |
| US20220375188A1 (en) | Semi-automatic image data labeling method, electronic apparatus, and storage medium | |
| CN109238292A (en) | A kind of Parcel division method and device thereof of unmanned machine operation | |
| CN109343701A (en) | A kind of intelligent human-machine interaction method based on dynamic hand gesture recognition | |
| JP2024528419A (en) | Method and apparatus for updating an object detection model | |
| CN114202026A (en) | Multitask model training method and device and multitask processing method and device | |
| CN113537070A (en) | Detection method, detection device, electronic equipment and storage medium | |
| GB2625664A (en) | Multi-task convolution-based object grasping and positioning identification algorithm and system, and robot | |
| CN116245193A (en) | Training method and device of target detection model, electronic equipment and medium | |
| CN109977884B (en) | Target following method and device | |
| CN110334818B (en) | Method and system for automatically identifying pipeline | |
| CN104376578A (en) | Moving object detection processing method and device applied to direct recording and broadcasting system | |
| CN113362227B (en) | Image processing method, device, electronic equipment and storage medium | |
| CN112132135B (en) | Power grid transmission line detection method based on image processing and storage medium | |
| CN110705695A (en) | Method, device, equipment and storage medium for searching model structure | |
| CN108957384A (en) | A kind of turnover box and electric energy measuring equipment discriminating direction method and apparatus | |
| CN115866229A (en) | Method, apparatus, device and medium for converting view angle of multi-view image | |
| CN112116561B (en) | Power grid transmission line detection method and device based on image processing fusion network weight | |
| CN111208785B (en) | Motion control method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160330 |
|
| RJ01 | Rejection of invention patent application after publication |