CN104392436A - Processing method and device for remote sensing image - Google Patents
Processing method and device for remote sensing image Download PDFInfo
- Publication number
- CN104392436A CN104392436A CN201410632672.4A CN201410632672A CN104392436A CN 104392436 A CN104392436 A CN 104392436A CN 201410632672 A CN201410632672 A CN 201410632672A CN 104392436 A CN104392436 A CN 104392436A
- Authority
- CN
- China
- Prior art keywords
- subimage
- module
- treatment facility
- sub
- resampling
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20021—Dividing image into blocks, subimages or windows
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30181—Earth observation
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
The embodiment of the invention discloses processing method and device for a remote sensing image. The method comprises the following steps: preprocessing a sub-image through a main processing device so as to lower down the data volume and processing complexity of the sub-image; determining whether a processing model in a processing model set of a secondary processing device meets the sub-image processing condition before processing the sub-image through the sub-processing device; if so, directly processing the sub-image through the processing model; if not so, calculating the required processing model again, and adding the re-calculated processing model to the processing model set so as to determine whether to meet the processing conditions of other sub-images. With the adoption of the processing method and device for the remote sensing image, the time an internal resource for re-calculating the processing model by the secondary processing device can be greatly saved; therefore, the internal resource and processing time of the processing device on the image can be reduced.
Description
Technical field
The present invention relates to remote sensing technology field, particularly relate to a kind of remote sensing image processing method and device.
Background technology
Remote sensing refers to contactless remote probing techniques, is widely used in the fields such as agricultural, forestry, geology, ocean, meteorology, the hydrology, military affairs, environmental protection.The remote sensing images obtained by remote sensing are signs of ground object, there is the feature of high information quantity, high data volume, the application of remote sensing images is mainly reflected in the extraction of remote sensing images information, the remote sensing images that sensor or other equipment directly obtain need to extract comprehensive, effective information after image procossing.
Remote sensing images also have ageing strong feature, such as, when the disaster such as earthquake, flood occurs, Real-time Obtaining disaster area remote sensing images, process remote sensing images and extract effective information and can grasp disaster area situation in the very first time as early as possible.For guaranteeing speed and the precision of remote sensing image processing, treatment facility needs to have higher hardware system configuration, and needs to distribute a large amount of internal resources to image processing work.Prior art is improve the processing speed of remote sensing images, and reduce the internal resource occupancy for the treatment of facility, the mode of general employing dispersion treatment, make the work of multiple treatment facility shared image procossing, namely be distributed to multiple sub-treatment facility after width remote sensing images are divided into multiple subimage in main treatment facility and carry out image procossing, sub-treatment facility is according to the process needs of pending subimage, transaction module required for computing subimage, and utilize this transaction module to process subimage, finally, subimage after process is sent back main treatment facility by each sub-treatment facility, and merge into the remote sensing images after a width process by main treatment facility.
But, when the remote sensing images that process data volume is larger, if the treatment facility negligible amounts of shared image processing work, the sub-image data amount then distributing to every platform treatment facility is still larger, and, the treatment facility transaction module calculated needed for each subimage needs to take a large amount of internal resources and operation time, there will be the problems such as the internal resource that image procossing takies treatment facility is comparatively large, the processing time is longer equally.
Summary of the invention
Provide a kind of remote sensing image processing method and device in the embodiment of the present invention, take to solve process remote sensing images the problem that treatment facility internal resource amount is comparatively large, the processing time is longer.
In order to solve the problems of the technologies described above, the embodiment of the invention discloses following technical scheme:
A kind of remote sensing image processing method, comprising:
Remote Sensing Image Segmentation is at least two subimages by main treatment facility;
Pre-service is carried out to described subimage;
Pretreated described subimage is distributed at least two sub-treatment facilities;
Described sub-treatment facility judges whether have the transaction module meeting described subimage in transaction module set, and described transaction module set is the set of transaction module described in described sub-treatment facility;
If described transaction module set has the transaction module meeting described subimage, the transaction module meeting described subimage is utilized to process described subimage;
If described transaction module set does not have the transaction module meeting described subimage, recalculate the transaction module meeting described subimage, and the transaction module that interpolation is recalculated enters described transaction module set; Utilization is recalculated the transaction module obtained and is processed described subimage;
Subimage after process is transferred to described main treatment facility by sub-treatment facility respectively described in each; Described main treatment facility merges the subimage after all process.
Alternatively, pre-service is carried out to described subimage, comprising:
Resampling process is carried out to described subimage, multiple pixels in preset range in described subimage are merged into a resampling pixel by described resampling process, carry out resampling process according to all pixels of preset order to described subimage, obtain the subimage after resampling;
Remove the noisy noise pixel point of tool in described resampling pixel;
The subimage after the described resampling of described noise pixel point is removed in compression.
Alternatively, the noisy noise pixel point of tool in described removal described resampling pixel, comprising:
If described resampling pixel meets noise conditions, the described resampling pixel meeting noise conditions is defined as described noise pixel point;
Utilize a predetermined number described resampling pixel adjacent with described noise pixel point to obtain denoising pixel, make described denoising pixel substitute described noise pixel point.
Alternatively, the subimage after the described resampling of described noise pixel point is removed in compression, comprising:
The first compressibility is adopted to compress to the first compression preset range in described subimage centered by described noise pixel point;
The second compressibility is adopted to compress to the second compression preset range in described subimage except described first compression preset range;
Described first compressibility is less than described second compressibility.
Alternatively, also comprise:
Described main treatment facility by segmentation after described subimage back-up storage in described main treatment facility, as the backup image of described subimage.
Alternatively, also comprise:
Described subimage after described main treatment facility combining data detection;
If the described subimage after merging has distortion compared with original described remote sensing images, obtain the described backup image corresponding with the described subimage with distortion;
Described sub-treatment facility described backup image is sent to process;
Backup image after process is transferred to described main treatment facility and substitutes the described subimage with distortion by described sub-treatment facility.
Alternatively, also comprise:
Adjacent described subimage has overlapped part at intersection.
Alternatively, also comprise:
Described main treatment facility obtains the maximum data treatment capacity of sub-treatment facility described in each, and splits described remote sensing images according to the described maximum data treatment capacity for the treatment of facility described in each.
A kind of remote sensing image processing device, is made up of main treatment facility and sub-treatment facility,
Described main treatment facility comprises segmentation module, pretreatment module, sending/receiving module and merging module, wherein,
It is at least two subimages that described segmentation module is used for described Remote Sensing Image Segmentation;
Described pretreatment module and described segmentation model calling, for carrying out pre-service to described subimage;
Described sending/receiving module and described pretreatment module and described merging model calling, for pretreated described subimage is distributed at least two described sub-treatment facilities, and the described subimage after described sub-treatment facility process is transferred to described merging module;
Described merging module is for merging the described subimage after all described sub-treatment facility process;
Described sub-treatment facility comprises model module and processing module, wherein,
Described model module meets the transaction module of described subimage in transaction module set for obtaining, when not there is the transaction module meeting described subimage in transaction module set, recalculate the transaction module meeting described subimage, and the transaction module that interpolation is recalculated enters described transaction module set, described transaction module set is the set of transaction module described in described sub-treatment facility;
Described processing module processes described subimage for utilizing the described transaction module meeting described subimage, and the described subimage after process is transferred to the described sending/receiving module of described main treatment facility.
Alternatively, described pretreatment module comprises:
Resampling module, for carrying out resampling process to described subimage, multiple pixels in preset range in described subimage are merged into a resampling pixel by described resampling process, carry out resampling process according to all pixels of preset order to described subimage, obtain the subimage after resampling;
Denoising module, for removing the noisy noise pixel point of tool in described resampling pixel;
Compression module, for remove described noise pixel point described resampling after subimage.
Alternatively, described denoising module comprises:
Obtain noise module, for when described resampling pixel meets noise conditions, the described resampling pixel meeting noise conditions is defined as described noise pixel point;
Removing noise module, for obtaining denoising pixel according to a predetermined number described resampling pixel adjacent with described noise pixel point, making described denoising pixel substitute described noise pixel point.
Alternatively, described compression module comprises:
First compression module, for adopting the first compressibility to compress to the first compression preset range in described subimage centered by described noise pixel point;
Second compression module, for adopting the second compressibility to compress to the second compression preset range in described subimage except described first compression preset range, described first compressibility is less than described second compressibility.
Alternatively, described main treatment facility also comprises:
Backup module, for by segmentation after described subimage back-up storage in described main treatment facility, as the backup image of described subimage.
Alternatively, described main treatment facility also comprises:
Detection module, for the described subimage after combining data detection, if the described subimage after merging has distortion compared with original described remote sensing images, obtains the described backup image corresponding with the described subimage with distortion;
Correct module, for described backup image being sent to described sub-treatment facility to process, and the described backup image after process is transferred to described main treatment facility and substitutes the described subimage with distortion.
Alternatively, described main treatment facility also comprises:
Obtaining information module, for obtaining the maximum data treatment capacity of sub-treatment facility described in each, described segmentation module splits described remote sensing images according to the maximum data treatment capacity of sub-treatment facility described in each in described acquisition module.
From above technical scheme, the remote sensing image processing method that the embodiment of the present invention provides and device, in main treatment facility, pre-service is carried out to subimage, reduce data volume and the process complexity of subimage, simultaneously, before sub-treatment facility processes subimage, first judge whether the transaction module set in sub-treatment facility has the transaction module that can meet subimage treatment conditions, if had, this transaction module is directly utilized to process subimage, if, do not recalculate required transaction module, and the transaction module recalculated is added in transaction module set, in order to judge whether the treatment conditions meeting other subimages, thus, greatly can save the internal resource that sub-treatment facility recalculates the time spent by transaction module and takies, thus, shared treatment facility internal resource and processing time during reduction image procossing, and, also the hardware configuration requirement for the treatment of facility can be reduced.
Accompanying drawing explanation
In order to be illustrated more clearly in the embodiment of the present invention or technical scheme of the prior art, be briefly described to the accompanying drawing used required in embodiment or description of the prior art below, apparently, for those of ordinary skills, under the prerequisite not paying creative work, other accompanying drawing can also be obtained according to these accompanying drawings.
The schematic flow sheet of a kind of remote sensing image processing method that Fig. 1 provides for the embodiment of the present invention;
The schematic flow sheet of the another kind of remote sensing image processing method that Fig. 2 provides for the embodiment of the present invention;
The structural representation of a kind of distant image processing apparatus that Fig. 3 provides for the embodiment of the present invention.
Embodiment
Technical scheme in the present invention is understood better in order to make those skilled in the art person, below in conjunction with the accompanying drawing in the embodiment of the present invention, technical scheme in the embodiment of the present invention is clearly and completely described, obviously, described embodiment is only the present invention's part embodiment, instead of whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art, not making the every other embodiment obtained under creative work prerequisite, should belong to the scope of protection of the invention.
Image reading apparatus etc. is utilized to change with the analog image that camera style obtains, obtaining general digital machine can reading number image, digital picture is made up of some pixels, and with ordinary numbers image similarity, the pixel of remote sensing images is little side's points with colourity and brightness.The image procossing of remote sensing images needs to carry out in image processing equipment, is made up of hardware (computing machine, display, digitizer, magnetic tape station etc.) and software (having the functions such as data input, output, correction, conversion, classification).
The schematic flow sheet of a kind of remote sensing image processing method that Fig. 1 provides for the embodiment of the present invention, be applied to by main treatment facility and sub-treatment facility co-treatment remote sensing images, main treatment facility in the embodiment of the present invention and sub-treatment facility can be computer, by internet transmission data and other information between main treatment facility and sub-treatment facility, as shown in Figure 1, comprise the following steps:
Step S101: main treatment facility segmentation remote sensing images;
Main treatment facility splits pending remote sensing images, it can be the subimage of identical or close size by remote sensing images even partition, or, also transversely with longitudinally two-dimentional all standing scanning can be carried out to remote sensing images, split according to the colourity of remote sensing images and the difference of brightness, by the weighted mean value determination split position of remote sensing images colourity and rate of change of brightness, wherein, the weighted value of colourity can be 70%, the weighted value of brightness can be 30%, scan whole remote sensing images, when weighted mean value rate of change exceedes preset value, mark a cut-point, each cut-point is connected to form cut-off rule, according to cut-off rule segmentation remote sensing images.
Step S102: pre-service is carried out to the subimage after segmentation;
Pre-service is carried out to the subimage after segmentation, image resampling, noise remove and compression etc. can be comprised.
Step S103: subimage is distributed to sub-treatment facility;
Pretreated subimage is distributed to sub-treatment facility, a width subimage can be sent to every sub-treatment facility, or, send to each sub-treatment facility the subimage that data volume varies in size according to the data-handling capacity of every the sub-treatment facility obtained in advance respectively.
Step S104: sub-treatment facility judges whether have the transaction module meeting subimage in transaction module set;
Transaction module is for the treatment of subimage, sub-treatment facility can calculate according to the processing requirements of subimage the transaction module meeting subimage, and utilize the transaction module obtained to process subimage, in the embodiment of the present invention, there is in sub-treatment facility the set of once calculated transaction module, set can be log file or the database of transaction module, when calculating the transaction module of subimage, first judge whether there is the transaction module meeting subimage in the transaction module set of sub-treatment facility, wherein, the transaction module meeting subimage is the transaction module meeting subimage treatment conditions, the treatment conditions of subimage can be the processing requirements of the subimage formulated according to different subimages and processing target, judge whether transaction module meets the processing requirements of subimage, transaction module that other subimages adjacent with pending subimage use when processing can be utilized to process pending subimage, if the subimage after process meets processing requirements, then determine that this transaction module meets the processing requirements of this pending subimage equally,
If had the transaction module meeting subimage in transaction module set, perform step S105; If do not have the transaction module meeting subimage in transaction module set, then perform step S106, sub-treatment facility recalculates the transaction module meeting subimage treatment conditions, then performs step S105 and step S107 simultaneously.
Step S105: utilize the transaction module meeting subimage to process subimage;
Utilize the pending subimage met in the transaction module antithetical phrase treatment facility of subimage to process, obtain the subimage after process.
Step S106: sub-treatment facility recalculates the transaction module meeting subimage;
Sub-treatment facility recalculates the transaction module meeting subimage according to the processing requirements of subimage.
Step S107: add the transaction module recalculated and enter transaction module set.
Subimage after process is transferred to main treatment facility by step S108: each sub-treatment facility respectively;
Subimage is after in sub-treatment facility, process terminates, and the subimage after process is sent back main treatment facility by subimage by each sub-treatment facility respectively.
Step S109: main treatment facility merges the subimage after all process;
Subimage after each process received merges by main treatment facility, is spliced into remote sensing images after a complete process.
Remote sensing images are split in main treatment facility, and after pre-service, be distributed to sub-treatment facility, reduce the requirement of sub-treatment facility process subimage for hardware configuration and internal memory, in addition, in actual process, adjacent subimage is due to the similarity between its image, the transaction module that needs during process adjacent sub-images may be made identical, therefore, before process subimage, first the transaction module whether having in sub-treatment facility and meet subimage treatment conditions is traveled through, if had, adopt this transaction module process subimage, without the need to calculating the transaction module of subimage.Because the time of computing model consumption is greatly more than the computing time utilizing transaction module process subimage to consume, so, adopt the method for the embodiment of the present invention can save the processing time of sub-treatment facility, also reduce the resource occupation amount of sub-treatment facility simultaneously.
In another embodiment of the present invention, the step S102 in above-described embodiment comprises the following steps:
21) resampling process is carried out to subimage, namely according to laterally order or longitudinal pixel merged sequentially successively in subimage in preset range of subimage, multiple pixels in preset range are permeated pixel, to reduce the size of data of subimage, in one particular embodiment of the present invention, preset range can comprise the neighbor pixel of more than 400, above-mentioned neighbor pixel can form the rectangle of a square or approximating square, neighbor pixel is merged into a resampling pixel according to selected function, after resampling process is carried out to the pixel in preset range, continue to carry out resampling process to all pixels in the next preset range adjacent with preset range in order, after successively resampling process being carried out to whole pixels of subimage, the all resampling pixels obtained form the subimage after resampling.
22) the noisy noise pixel point of tool in resampling pixel is removed;
Remote sensing images are in the process obtained, inevitably produce some noise spots, the similarity that noise spot shows as between neighbor is poor, be generally the pixel of high brightness, subimage is through step 21) after resampling, the noisy noise pixel point of tool in the subimage after utilizing image de-noising method to remove resampling.
23) subimage after the resampling of removing noise pixel point is compressed;
Perform step 22) after, the noise spot of subimage is removed, and carries out compression process, reduce the data volume of subimage further to the subimage after denoising.
Not only reduce data volume by pretreated subimage, also reduce the complexity of sub-treatment facility process subimage simultaneously, thus reduce processing time and the internal resource occupancy of sub-treatment facility.
In another embodiment of the present invention, the step 22 in above-described embodiment) comprise the following steps:
221) chromatic value and the brightness value of each resampling pixel in subimage is after pretreatment obtained, the chromatic value of resampling pixel and brightness value are compared with normal distribution curve, if certain resampling pixel is positioned at outside normal distribution curve, then this resampling pixel meets noise conditions, remove this resampling pixel, wherein, normal distribution curve is:
In formula, parameter μ is 0, parameter σ is 1.5, and in embodiments of the present invention, parameters can be selected according to the actual needs of image procossing.
222) the some resampling pixels adjacent with being removed pixel are obtained, namely the adjacent resampling pixel of noise pixel point is obtained, such as, obtain 8 resampling pixels around noise pixel point, permeate some adjacent resampling pixels a pixel, pixel after fusion is filled up in the position being removed pixel, completes noise removal process.
In another specific embodiment of the present invention, the mode of wavelet transformation can also be adopted to remove noise in image;
Small echo is utilized to carry out denoising to noisy image, set suitable threshold value, first will be less than the coefficient zero of threshold value, retain the coefficient being greater than threshold value, and then map through threshold function table and obtain estimation coefficient, the denoising of image just can be realized through inverse transformation, wherein, threshold value is (1+x2)/(1-k*e-x), x is the signal to noise ratio (S/N ratio) of wavelet packet, k is scale-up factor, and k can be 0.2.
In another embodiment of the present invention, above-mentioned steps 23) comprise the following steps:
231) the first compressibility is adopted to compress to the first compression preset range in subimage centered by noise pixel point;
In the present invention's specific embodiment, preset range can for noise pixel point for the center of circle and area accounts for the circular portion of subimage 0.01%, adopt the first compressibility to compress to this circular portion.
232) adopt the second compressibility to compress to except the first the second compression preset range compressed except preset range being the center of circle with noise pixel point in subimage, the first compressibility is less than the second compressibility.
Adopt above-mentioned compress mode, make the compression degree of noise pixel point near zone lower than other regions, can avoid after noise remove pixel due to compression degree excessive and the situation of distortion occurs.
In another embodiment of the present invention, as shown in Figure 2, after performing the step S101 in above-described embodiment, step S201 is performed,
Step S201: backup subimage;
By the subimage back-up storage after segmentation in main treatment facility, as the backup image of subimage, and set up the index between backup image and subimage; The database of storage backup image can adopt stack architecture, after the process completing view picture remote sensing images, delete the backup image stored in main treatment facility, avoid unnecessary the taking of device interior resource, for when processing remote sensing images, storage backup image is prepared next time.
In execution above-described embodiment after step S109, perform step S202,
Step S202: judge whether the subimage after merging has distortion compared with original remote sensing images;
Because remote sensing images are after a series of images process, the distortion such as anamorphose, inclination may be there is, therefore, subimage after combining data detection, if the subimage after merging has distortion compared with original remote sensing images, perform step S203, in main treatment facility, obtain the backup image with the subimage of distortion; If the subimage after merging does not distort compared with original remote sensing images, perform step S204, complete the process to remote sensing images.
Step S203: obtain the backup image with the subimage of distortion;
In the subimage backed up in advance, according to the backup image that index search is corresponding with the subimage with distortion, and this backup image is sent to sub-treatment facility, perform the step S103 in above-described embodiment and other steps follow-up, until there is not the subimage distorted after obtaining process, the subimage that utilizes this not distort replaces the former subimage with distortion, merges all subimages meeting processing requirements, forms the remote sensing images after a complete process.
By backup subimage, the subimage handled by sub-treatment facility, once generation distortion, can utilize the subimage of backup to re-start process, to ensure that final acquisition meets the remote sensing images of processing requirements.
In another embodiment of the present invention, adjacent subimage has overlapping part at intersection, can be overlapped, the width of overlapping region can be the width of 300 pixels, to ensure that each subimage is when final splicing merges, better excessive effects can be realized by overlapping region, avoid splicing improperly to make whole remote sensing images occur distortion.
In another embodiment of the present invention, before performing above-mentioned steps S101, following step is first performed: main treatment facility obtains the maximum data treatment capacity of each sub-treatment facility;
Main treatment facility collects the maximum data treatment capacity that each sub-treatment facility feeds back respectively, on the processing power basis knowing each sub-treatment facility, perform the step S101 in above-described embodiment, according to the maximum data treatment capacity segmentation remote sensing images of individual sub-treatment facility, such as, have 3 sub-treatment facility co-treatment one width remote sensing images, then by the processing power cutting remote sensing images of remote sensing images according to sub-treatment facility, the data volume of every width subimage is made all to be less than the maximum data treatment capacity of distributed sub-treatment facility, to ensure that every sub-treatment facility all can process subimage under normal processing speed.
A kind of remote sensing image processing device that Fig. 3 provides for the embodiment of the present invention, is made up of main treatment facility 1 and sub-treatment facility 2, and wherein main treatment facility 1 comprises segmentation module 3, pretreatment module 4, merges module 5 and sending/receiving module 6; Sub-treatment facility 2 comprises model module 7 and processing module 8;
Segmentation module 3 in main treatment facility 1 is for being at least two subimages by Remote Sensing Image Segmentation;
Pretreatment module 4 in main treatment facility 1 is connected, for carrying out pre-service to subimage with segmentation module 3;
Sending/receiving module 6 in main treatment facility 1 and pretreatment module 4 and merges module 5 and be connected, for pretreated subimage being distributed at least two sub-treatment facilities 2, and the subimage after being processed by sub-treatment facility 2 is transferred to merging module 5;
Merging module 5 in main treatment facility 1 is for merging the subimage after the process of all sub-treatment facilities 2;
Model module 7 in sub-treatment facility 2 is for obtaining the transaction module meeting subimage in transaction module set, when not there is the transaction module meeting subimage in transaction module set, recalculate the transaction module meeting subimage, and the transaction module that interpolation is recalculated enters transaction module set, transaction module set is the set of transaction module in sub-treatment facility;
Processing module 8 in sub-treatment facility 2 processes subimage for the transaction module meeting subimage utilizing model module 7 to obtain, and the subimage after process is transferred to the sending/receiving module 6 of main treatment facility 1.
In another embodiment of the present invention, the pretreatment module 4 in above-described embodiment comprises:
Resampling module, for carrying out resampling process to subimage, multiple pixels in preset range in subimage are merged into a resampling pixel by resampling process, resampling process is carried out according to all pixels of preset order to subimage, obtain the subimage after resampling, such as, carry out resampling along the horizontal direction of subimage or longitudinal direction;
Denoising module, for removing the noisy noise pixel point of tool in resampling pixel;
Compression module, for compressing the subimage after removing noise pixel point.
In another embodiment of the present invention, the denoising module in above-described embodiment comprises:
Obtain noise module, for when resampling pixel meets noise conditions, the resampling pixel meeting noise conditions is defined as noise pixel point;
Removing noise module, for obtaining denoising pixel according to the resampling pixel adjacent with noise pixel point, utilizing denoising pixel to substitute noise pixel point.
In another embodiment of the present invention, the compression module in above-described embodiment comprises:
First compression module, for adopting the first compressibility to compress to the first compression preset range in subimage centered by noise pixel point;
Second compression module, for adopting the second compressibility to compress to the second compression preset range in subimage except the first compression preset range, the first compressibility is less than the second compressibility.
In another embodiment of the present invention, the main treatment facility 1 in above-described embodiment also comprises:
Backup module, for by segmentation after subimage back-up storage in main treatment facility 1, as the backup image of subimage.
In another embodiment of the present invention, the main treatment facility 1 in above-described embodiment also comprises:
Detection module, for the subimage after combining data detection, if the subimage after merging has distortion compared with original remote sensing images, obtains the backup image corresponding with the subimage with distortion;
Correct module, for backup image being sent to sub-treatment facility to process, and the backup image after process is transferred to main treatment facility and substitutes the subimage with distortion.
In another embodiment of the present invention, the main treatment facility 1 in above-described embodiment also comprises:
Obtaining information module, for obtaining the maximum data treatment capacity of each sub-treatment facility, segmentation module 3 is according to the maximum data treatment capacity segmentation remote sensing images of each sub-treatment facility in acquisition module.
It should be noted that, in this article, the such as relational terms of " first " and " second " etc. and so on is only used for an entity or operation to separate with another entity or operational zone, and not necessarily requires or imply the relation that there is any this reality between these entities or operation or sequentially.And, term " comprises ", " comprising " or its any other variant are intended to contain comprising of nonexcludability, thus make to comprise the process of a series of key element, method, article or equipment and not only comprise those key elements, but also comprise other key elements clearly do not listed, or also comprise by the intrinsic key element of this process, method, article or equipment.When not more restrictions, the key element limited by statement " comprising ... ", and be not precluded within process, method, article or the equipment comprising described key element and also there is other identical element.
The above is only the specific embodiment of the present invention, those skilled in the art is understood or realizes the present invention.To be apparent to one skilled in the art to the multiple amendment of these embodiments, General Principle as defined herein can without departing from the spirit or scope of the present invention, realize in other embodiments.Therefore, the present invention can not be restricted to these embodiments shown in this article, but will meet the widest scope consistent with principle disclosed herein and features of novelty.
Claims (15)
1. a remote sensing image processing method, is characterized in that, comprising:
Remote Sensing Image Segmentation is at least two subimages by main treatment facility;
Pre-service is carried out to described subimage;
Pretreated described subimage is distributed at least two sub-treatment facilities;
Described sub-treatment facility judges whether have the transaction module meeting described subimage in transaction module set, and described transaction module set is the set of transaction module described in described sub-treatment facility;
If described transaction module set has the transaction module meeting described subimage, the transaction module meeting described subimage is utilized to process described subimage;
If described transaction module set does not have the transaction module meeting described subimage, recalculate the transaction module meeting described subimage, and the transaction module that interpolation is recalculated enters described transaction module set; Utilization is recalculated the transaction module obtained and is processed described subimage;
Subimage after process is transferred to described main treatment facility by sub-treatment facility respectively described in each; Described main treatment facility merges the subimage after all process.
2. method according to claim 1, is characterized in that, carries out pre-service, comprising described subimage:
Resampling process is carried out to described subimage, multiple pixels in preset range in described subimage are merged into a resampling pixel by described resampling process, carry out resampling process according to all pixels of preset order to described subimage, obtain the subimage after resampling;
Remove the noisy noise pixel point of tool in described resampling pixel;
The subimage after the described resampling of described noise pixel point is removed in compression.
3. method according to claim 2, is characterized in that, the noisy noise pixel point of tool in described removal described resampling pixel, comprising:
If described resampling pixel meets noise conditions, the described resampling pixel meeting noise conditions is defined as described noise pixel point;
Utilize a predetermined number described resampling pixel adjacent with described noise pixel point to obtain denoising pixel, make described denoising pixel substitute described noise pixel point.
4. method according to claim 2, is characterized in that, the subimage after the described resampling of described noise pixel point is removed in described compression, comprising:
The first compressibility is adopted to compress to the first compression preset range in described subimage centered by described noise pixel point;
The second compressibility is adopted to compress to the second compression preset range in described subimage except described first compression preset range;
Described first compressibility is less than described second compressibility.
5. method according to claim 1, is characterized in that, also comprises:
Described main treatment facility by segmentation after described subimage back-up storage in described main treatment facility, as the backup image of described subimage.
6. method according to claim 5, is characterized in that, also comprises:
Described subimage after described main treatment facility combining data detection;
If the described subimage after merging has distortion compared with original described remote sensing images, obtain the described backup image corresponding with the described subimage with distortion;
Described sub-treatment facility described backup image is sent to process;
Backup image after process is transferred to described main treatment facility and substitutes the described subimage with distortion by described sub-treatment facility.
7. method according to claim 1, is characterized in that, also comprises:
Adjacent described subimage has overlapped part at intersection.
8. method according to claim 1, is characterized in that, also comprises:
Described main treatment facility obtains the maximum data treatment capacity of sub-treatment facility described in each, and splits described remote sensing images according to the described maximum data treatment capacity for the treatment of facility described in each.
9. a remote sensing image processing device, is made up of main treatment facility and sub-treatment facility, it is characterized in that,
Described main treatment facility comprises segmentation module, pretreatment module, sending/receiving module and merging module, wherein,
It is at least two subimages that described segmentation module is used for described Remote Sensing Image Segmentation;
Described pretreatment module and described segmentation model calling, for carrying out pre-service to described subimage;
Described sending/receiving module and described pretreatment module and described merging model calling, for pretreated described subimage is distributed at least two described sub-treatment facilities, and the described subimage after described sub-treatment facility process is transferred to described merging module;
Described merging module is for merging the described subimage after all described sub-treatment facility process;
Described sub-treatment facility comprises the model module and processing module that are connected, wherein,
Described model module meets the transaction module of described subimage in transaction module set for obtaining, when not there is the transaction module meeting described subimage in transaction module set, recalculate the transaction module meeting described subimage, and the transaction module that interpolation is recalculated enters described transaction module set, described transaction module set is the set of transaction module described in described sub-treatment facility;
Described processing module processes described subimage for utilizing the described transaction module meeting described subimage, and the described subimage after process is transferred to the described sending/receiving module of described main treatment facility.
10. device according to claim 9, is characterized in that, described pretreatment module comprises:
Resampling module, for carrying out resampling process to described subimage, multiple pixels in preset range in described subimage are merged into a resampling pixel by described resampling process, carry out resampling process according to all pixels of preset order to described subimage, obtain the subimage after resampling;
Denoising module, for removing the noisy noise pixel point of tool in described resampling pixel;
Compression module, for remove described noise pixel point described resampling after subimage.
11. devices according to claim 10, is characterized in that, described denoising module comprises:
Obtain noise module, for when described resampling pixel meets noise conditions, the described resampling pixel meeting noise conditions is defined as described noise pixel point;
Removing noise module, for obtaining denoising pixel according to a predetermined number described resampling pixel adjacent with described noise pixel point, making described denoising pixel substitute described noise pixel point.
12. devices according to claim 10, is characterized in that, described compression module comprises:
First compression module, for adopting the first compressibility to compress to the first compression preset range in described subimage centered by described noise pixel point;
Second compression module, for adopting the second compressibility to compress to the second compression preset range in described subimage except described first compression preset range, described first compressibility is less than described second compressibility.
13. devices according to claim 9, is characterized in that, described main treatment facility also comprises:
Backup module, for by segmentation after described subimage back-up storage in described main treatment facility, as the backup image of described subimage.
14. devices according to claim 13, is characterized in that, described main treatment facility also comprises:
Detection module, for the described subimage after combining data detection, if the described subimage after merging has distortion compared with original described remote sensing images, obtains the described backup image corresponding with the described subimage with distortion;
Correct module, for described backup image being sent to described sub-treatment facility to process, and the described backup image after process is transferred to described main treatment facility and substitutes the described subimage with distortion.
15. devices according to claim 9, is characterized in that, described main treatment facility also comprises:
Obtaining information module, for obtaining the maximum data treatment capacity of sub-treatment facility described in each, described segmentation module splits described remote sensing images according to the maximum data treatment capacity of sub-treatment facility described in each in described acquisition module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410632672.4A CN104392436B (en) | 2014-11-11 | 2014-11-11 | A kind of remote sensing image processing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410632672.4A CN104392436B (en) | 2014-11-11 | 2014-11-11 | A kind of remote sensing image processing method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104392436A true CN104392436A (en) | 2015-03-04 |
| CN104392436B CN104392436B (en) | 2019-01-04 |
Family
ID=52610334
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410632672.4A Active CN104392436B (en) | 2014-11-11 | 2014-11-11 | A kind of remote sensing image processing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104392436B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106557284A (en) * | 2016-12-08 | 2017-04-05 | 天津汉铭科技发展有限公司 | Remote sensing image processing method and device |
| CN108416353A (en) * | 2018-02-03 | 2018-08-17 | 华中农业大学 | Crop field spike of rice fast partition method based on the full convolutional neural networks of depth |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101706950A (en) * | 2009-07-13 | 2010-05-12 | 中国科学院遥感应用研究所 | High-performance implementation method for multi-scale segmentation of remote sensing images |
| CN102289792A (en) * | 2011-05-03 | 2011-12-21 | 北京云加速信息技术有限公司 | Method and system for enhancing low-illumination video image |
| CN102438134A (en) * | 2011-11-22 | 2012-05-02 | 苏州科雷芯电子科技有限公司 | Video transmission apparatus based on visual center principle and transmission method thereof |
| CN103180878A (en) * | 2010-10-25 | 2013-06-26 | 皇家飞利浦电子股份有限公司 | System for the segmentation of a medical image |
| US8620082B1 (en) * | 2011-04-01 | 2013-12-31 | The United States Of America As Represented By The Secretary Of The Navy | Sonar image texture segmentation |
| CN103778413A (en) * | 2014-01-16 | 2014-05-07 | 华东师范大学 | Remote-sensing image under-segmentation object automatic recognition method |
-
2014
- 2014-11-11 CN CN201410632672.4A patent/CN104392436B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101706950A (en) * | 2009-07-13 | 2010-05-12 | 中国科学院遥感应用研究所 | High-performance implementation method for multi-scale segmentation of remote sensing images |
| CN103180878A (en) * | 2010-10-25 | 2013-06-26 | 皇家飞利浦电子股份有限公司 | System for the segmentation of a medical image |
| US8620082B1 (en) * | 2011-04-01 | 2013-12-31 | The United States Of America As Represented By The Secretary Of The Navy | Sonar image texture segmentation |
| CN102289792A (en) * | 2011-05-03 | 2011-12-21 | 北京云加速信息技术有限公司 | Method and system for enhancing low-illumination video image |
| CN102438134A (en) * | 2011-11-22 | 2012-05-02 | 苏州科雷芯电子科技有限公司 | Video transmission apparatus based on visual center principle and transmission method thereof |
| CN103778413A (en) * | 2014-01-16 | 2014-05-07 | 华东师范大学 | Remote-sensing image under-segmentation object automatic recognition method |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106557284A (en) * | 2016-12-08 | 2017-04-05 | 天津汉铭科技发展有限公司 | Remote sensing image processing method and device |
| CN108416353A (en) * | 2018-02-03 | 2018-08-17 | 华中农业大学 | Crop field spike of rice fast partition method based on the full convolutional neural networks of depth |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104392436B (en) | 2019-01-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102800052B (en) | Semi-automatic digital method of non-standard map | |
| CN109712071B (en) | Unmanned aerial vehicle image splicing and positioning method based on track constraint | |
| CN107480203A (en) | It is a kind of to be directed to identical and similar pictures duplicate removal view data cleaning method | |
| CN110942071A (en) | License plate recognition method based on license plate classification and LSTM | |
| US8249387B2 (en) | Image processing method and apparatus for detecting lines of images and start and end points of lines | |
| CN110599455A (en) | Display screen defect detection network model, method and device, electronic equipment and storage medium | |
| CN115115611B (en) | Vehicle damage identification method and device, electronic equipment and storage medium | |
| CN113421242A (en) | Deep learning-based welding spot appearance quality detection method and device and terminal | |
| CN112883900A (en) | Method and device for bare-ground inversion of visible images of remote sensing images | |
| CN110008949B (en) | Image target detection method, system, device and storage medium | |
| CN104392436A (en) | Processing method and device for remote sensing image | |
| CN114359932A (en) | Text detection method, text recognition method and text recognition device | |
| CN117132737B (en) | Three-dimensional building model construction method, system and equipment | |
| CN115880288B (en) | Detection method, system and computer equipment for electronic element welding | |
| CN112270259A (en) | SAR image ship target rapid detection method based on lightweight convolutional neural network | |
| CN102778684A (en) | Embedded monocular passive target tracking positioning system and method based on FPGA (Field Programmable Gate Array) | |
| CN116597270A (en) | Road damage target detection method based on attention mechanism integrated learning network | |
| CN104318236A (en) | Method and system for obtaining image local features | |
| CN114419428A (en) | Target detection method, target detection device and computer readable storage medium | |
| CN114373078A (en) | Target detection method and device, terminal equipment and storage medium | |
| CN113643173A (en) | Watermark removing method, watermark removing device, terminal equipment and readable storage medium | |
| CN113610856A (en) | Method and device for training image segmentation model and image segmentation | |
| CN104408437A (en) | Road detecting method based on synthetic aperture radar | |
| CN114973910B (en) | Map generation method and device, electronic equipment and computer readable medium | |
| CN115527074B (en) | Vehicle detection frame generation method and device 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |