CN104537622A - Method and system for removing raindrop influence in single image - Google Patents
Method and system for removing raindrop influence in single image Download PDFInfo
- Publication number
- CN104537622A CN104537622A CN201410856519.XA CN201410856519A CN104537622A CN 104537622 A CN104537622 A CN 104537622A CN 201410856519 A CN201410856519 A CN 201410856519A CN 104537622 A CN104537622 A CN 104537622A
- Authority
- CN
- China
- Prior art keywords
- image
- rain
- remove
- hfs
- pending image
- 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
Landscapes
- Image Analysis (AREA)
Abstract
The invention provides a method and system for removing the raindrop influence in a single image. The method includes the steps that the image to be processed is decomposed based on an empirical mode decomposition method, the high-frequency portion of the image to be processed is extracted, and a high-frequency feature map reflecting information of the high-frequency portion is formed; the edges of image elements in the image to be processed are recognized, and a feature contour map is acquired; an image region in the edges is processed through image morphological operation, and a first intermediate image is acquired; a rain streak feature map is acquired by subtracting the first intermediate image from the high-frequency feature map; the image with the raindrop influence removed is acquired by subtracting the rain streak feature map from the image to be processed. The image processing speed is effectively increased through the method and system.
Description
Technical field
The present invention relates to digital image processing techniques, particularly relate in a kind of single image the method and system removing raindrop impact.
Background technology
Comprise the plurality of advantages such as automatism, intelligent, high efficiency owing to having, outdoor computer vision system is widely used in the fields such as military and national defense, medical skill, intelligent transportation.But inclement weather can have a strong impact on its performance, even cause its complete failure.So eliminate the effective ways of adverse weather conditions, for essential a round-the-clock outdoor vision system.In many inclement weather conditions, rain, owing to having comparatively macroparticle (raindrop) radius and other complicated physical characteristicss, can cause the quality of the image that vision system absorbs and affect largely.Image raindrop remove technology by using the characteristic such as physics, frequency of rain, identify the raindrop in image, remove.It significantly can not only promote picture quality, also helps the further process of image.Therefore, image raindrop remove technology has become the indispensable guardian technique of computer vision field.
Detected about raindrop in image in the last few years and became focus already with the research of removing.Starik etc. proposed time domain average the earliest raindrop in 2003 remove strategy, author thinks in sequence of video images, raindrop are only present in several frame the impact of pixel, therefore directly can be averaged on frame of video and just can obtain the original image of the impact eliminating rain.Regrettably, they do not carry out verification experimental verification to method.Garg and Nayar employs the dynamic of rain and photometric property (K.Garg and S.K.Nayar the earliest, " Detection and removalof rain from videos; " in Proc.IEEE Conf.Comput.Vis.Pattern Recognit., Jun.2004, vol.1, pp.528 – 535), establish two kinds of models respectively, and propose the method detecting and remove rain based on these two models.For the dynamic model of rain, which show rain, at its falling direction, there is relativity of time domain; For luminosity model, it is divided into static rain and dynamic rain model.For static raindrop, its brightness is significantly higher than its background covered; For dynamic raindrop (rain line), its brightness is by static raindrop brightness, background luminance and camera exposure Time dependent.Afterwards, author proposes a kind of frame difference method that uses and carries out raindrop initial survey, uses two kinds of characteristics to carry out flase drop removal, and the final method utilizing front and back frame image information to carry out raindrop removal.Although the method better performances, the rain in its rain for seriously (at a distance) out of focus, bright background and force of rain change cannot process.People (the Zhang X P such as Zhang in 2006, Li H, Qi Y Y, Leow W K, Ng T K.Rainremoval in video by combining temporal and chromatic properties.In:Proceedings ofthe 2006International Conferenceon Multimedia and Expo.Toronto, Canada:IEEE, 2006.461:464) employ rain time domain distribution and chromatic characteristic.Time domain distribution histogram due to rain shows two peaks (representing raindrop brightness and background luminance respectively), and approximate formation gauss hybrid models, therefore unsupervised learning method---K-means cluster can be separated it effectively.Afterwards, author finds that the change being affected the interframe rgb value of pixel by raindrop is substantially identical, therefore flase drop can be removed further.The method experiment effect is better, but utilizes the method for cluster to distinguish raindrop and background at whole video, and counting yield is not high, can not carry out real-time process.People (the Barnum P C such as Barnum in 2007, Narasimhan S G, KanadeT.Analysis of rainand snow in frequency space.Internatio-nal Journal of ComputerVision, 2010,86 (2:3): 256:274) notice before most methods depend critically upon the extraction of clear rain line, and rain line is owing to can cause the pattern of repetition, it is rational for carrying out analysis to rain in a frequency domain.Author sets up the impact that Gauss model carrys out approximate rain, and by asking the model proportion in three-dimensional Fourier transform to carry out raindrop detection, and then remove rain by iteration, last inverse transformation is to video image.Experimental result shows that this kind of method has good handling property, but the time complexity of the method is too high, and for the process that inconspicuous rain and the force of rain change, it there will be remarkable hydraulic performance decline.
Above go rain method based on single image, how only gray level image can be processed, and method required time is longer, such as up-to-date optimized algorithm (methods of Chen etc.), process the time of specific single image at more than 100s, simultaneously to there will be to a certain extent fuzzy for output image.
Remove the time complexity of technology based on single image raindrop in prior art too high, be unfavorable for the shortcoming of the popularization of method, need the raindrop removal technology improved further in image.
Summary of the invention
Based on this, be necessary for prior art Problems existing, provide in a kind of single image the method and system removing raindrop impact, it can process coloured image, and effectively improves the processing speed of image.
Remove a method for raindrop impact in single image, it comprises:
Based on empirical mode decomposition method, picture breakdown is carried out to pending image, extract the HFS of described pending image, form the high-frequency characteristic figure embodying described HFS information;
Identify the edge of pictorial element in described pending image, obtain feature contour figure;
Utilize morphological image to operate to process described intramarginal image-region, obtain the first intermediate image;
From described high-frequency characteristic figure, deduct described first intermediate image, obtain rain line features figure;
Described pending image and described rain line features figure are subtracted each other, obtains the image after removing rain.
Wherein in an embodiment, described method also comprises:
Described remove rain after image on superpose described feature contour figure, obtain repair after image.
The process of the HFS wherein in an embodiment, describedly carry out picture breakdown based on empirical mode decomposition method to pending image, extracting described pending image comprises the following steps:
By described pending image mapped to XOY plane, the gray-scale value of described pending image respective pixel is as Z coordinate;
Identified local maximum and the minimal value of described pending image by morphological method, obtain multiple scattered maximum point and minimum point;
Described multiple scattered maximum point and minimum point are carried out respectively to the triangulation of plane point set, then interpolation smoothing obtains maximum value envelope surface and minimal value envelope surface;
Calculate the average of described maximum value envelope surface and minimal value envelope surface;
The gray-scale value of each pixel in described pending image is deducted described average, obtains exploded view picture;
Judge that described exploded view similarly is no satisfied screening termination condition, if so, then described exploded view picture is exported as described HFS; If not, then return described in execution and identify the local maximum of described pending image and minimizing step by morphological method.
Wherein in an embodiment, described described exploded view picture also to be comprised before the step that described HFS exports:
From described pending image, deduct described exploded view picture, obtain when the image after pre-treatment;
When whether the image after pre-treatment meets picture breakdown termination condition, if then exported as described HFS by described exploded view picture described in judging;
If not, then decompose next time, return to perform and identify the local maximum of described pending image and minimizing step by morphological method, until meet described picture breakdown termination condition, export the described exploded view picture repeatedly decomposing acquisition, export as described HFS.
Wherein in an embodiment, in the described pending image of described identification, the edge of pictorial element obtains in the process of feature contour figure and adopts Privett (Prewitt) operator to carry out rim detection to image based on gradation of image.
Wherein in an embodiment, described utilize morphological image operate the process that described intramarginal image-region processes is comprised: fill based on the Contiguous graphics region of image procossing etching operation to described edge.
Wherein in an embodiment, describedly from described high-frequency characteristic figure, deduct the process that described first intermediate image obtains rain line features figure comprise:
To described first intermediate image negate, obtain the second intermediate image;
Be multiplied with described high-frequency characteristic figure with described second intermediate image, extract the common factor of described second intermediate image and described high-frequency characteristic figure, form described rain line features figure.
Wherein in an embodiment, described described pending image and described rain line features figure are subtracted each other to obtain go the process of the image after rain to comprise:
To described rain line features figure negate, obtain the 3rd intermediate image;
Be multiplied with described pending image with described 3rd intermediate image, extract the common factor of described 3rd intermediate image and described pending image, described in formation, remove the image after rain.
Remove a system for raindrop impact in single image, it comprises:
Picture breakdown module, for carrying out picture breakdown based on empirical mode decomposition method to pending image, extracts the HFS of described pending image, forms the high-frequency characteristic figure embodying described HFS information;
Edge detection module, for identifying the edge of pictorial element in described pending image, obtains feature contour figure;
Packing module, operates for utilizing morphological image and processes described intramarginal image-region, obtain the first intermediate image;
First computing module, for deducting described first intermediate image from described high-frequency characteristic figure, obtains rain line features figure; And
Second computing module, for described pending image and described rain line features figure being subtracted each other, obtains the image after removing rain.
Wherein in an embodiment, described system also comprises:
Overlap-add procedure module, for described remove rain after image on superpose described feature contour figure, obtain repair after image.
Based on said method and system, the present invention, by the picture breakdown technology based on empirical mode decomposition, obtains the HFS of image, re-uses image border recognizer, both subtract each other at result, thus finally obtain the pixel of being rung by rain shadow, because the pixel intensity of being rung by rain shadow is higher, deduct it finally by from former figure, obtain rain image, the present invention effectively can improve the visual effect of ringing image by rain shadow, can process coloured image, and improve arithmetic speed.Utilize method of the present invention the single image processing time can be reduced about 50%.
Accompanying drawing explanation
Fig. 1 is the schematic flow sheet of an embodiment of the inventive method;
Fig. 2 is the schematic flow sheet of another embodiment of the inventive method;
Fig. 3 is the structural representation of an embodiment of present system;
Fig. 4 is the design sketch of pending image;
Fig. 5 is the design sketch of the image after the reparation in one embodiment of the invention.
Embodiment
Image based on field of machine vision goes rain technology, the present invention proposes a kind of new image and go rain method, it is by the picture breakdown technology based on empirical mode decomposition, obtain the HFS of image, re-use image border recognizer, both subtract each other at result, thus finally obtain the pixel of being rung by rain shadow, because the pixel intensity of being rung by rain shadow is higher, deduct it finally by from former figure, obtain rain image, the present invention effectively can improve the visual effect of ringing image by rain shadow, the computation process of simplified image process, accelerates image processing efficiency.Each embodiment of the present invention is described in detail below with reference to accompanying drawing.
As shown in Figure 1, the invention provides in a kind of single image the method removing raindrop impact, it comprises the following steps.
In step 100, based on empirical mode decomposition method, picture breakdown is carried out to pending image, extract the HFS of above-mentioned pending image, form the high-frequency characteristic figure embodying above-mentioned HFS information.Here the empirical mode decomposition method mentioned refers to a kind of new Non-stationary Signal Analysis method, has the advantage such as locality, adaptivity.In one embodiment of the invention, Bidimensional Empirical Mode Decomposition (BEMD, Bidimensional Empirical Mode Decomposition) method is adopted to carry out picture breakdown to pending image.And for example, in a preferred embodiment of the invention, as shown in Figure 2, the implementation of above-mentioned steps 100 comprises the following steps:
Step 101, by above-mentioned pending image mapped to XOY plane, the gray-scale value of above-mentioned pending image respective pixel is as Z coordinate;
Step 102, is identified local maximum and the minimal value of above-mentioned pending image, obtains multiple scattered maximum point and minimum point by morphological method;
Step 103, above-mentioned multiple scattered maximum point and minimum point are carried out respectively to the triangulation of plane point set, then interpolation smoothing obtains maximum value envelope surface and minimal value envelope surface;
Step 104, calculates the average of above-mentioned maximum value envelope surface and minimal value envelope surface;
Step 105, deducts above-mentioned average by the gray-scale value of each pixel in above-mentioned pending image, obtains exploded view picture;
Step 106, judge that above-mentioned exploded view similarly is no satisfied screening termination condition, this screening termination condition is specially the inspection of zero crossing condition and average condition.If extreme point (this extreme point comprises maximum point and minimum point) number is equal with across numbers of zeros or maximum difference one and the above-mentioned average that is made up of local maximum are zero, then perform step 109, if not, then return execution above-mentioned steps 102 and identify the local maximum of above-mentioned pending image and minimizing step by morphological method.
Step 109, using the image detail information that above-mentioned exploded view picture obtains as this decomposable process, if above-mentioned exploded view picture meets screening termination condition, above-mentioned HFS can be used as and export, multiple image detail informations that decomposable process obtains if exist repeatedly, then superpose the above-mentioned exploded view picture repeatedly decomposing acquisition as above-mentioned high-frequency characteristic figure.And for example, in another embodiment of the present invention, above-mentioned, above-mentioned exploded view picture was also comprised before the step 109 that above-mentioned HFS exports:
Step 107, deducts above-mentioned exploded view picture from above-mentioned pending image, obtains when the image after pre-treatment;
Step 108, judge above-mentionedly whether meet picture breakdown termination condition when the image after pre-treatment, whether this picture breakdown termination condition is specially every tomographic image detailed information has and is no more than an extreme point, if then perform above-mentioned steps 109: exported as above-mentioned HFS by above-mentioned exploded view picture;
If not, then iterations adds one in order to decompose next time, and return perform above-mentioned steps 102 identify the local maximum of above-mentioned pending image and minimizing step by morphological method, until meet above-mentioned picture breakdown termination condition, export and repeatedly decompose the above-mentioned exploded view picture obtained respectively, export as above-mentioned HFS.
Further, in one embodiment of the invention, the step that above-mentioned formation embodies the high-frequency characteristic figure of HFS information is: if above-mentioned HFS only comprises the exploded view picture that decomposable process obtains, then using this exploded view picture as above-mentioned high-frequency characteristic figure; If above-mentioned HFS comprises the exploded view picture that repeatedly decomposable process obtains respectively, then superpose the above-mentioned exploded view picture repeatedly decomposing acquisition, form above-mentioned high-frequency characteristic figure.
Utilize above-mentioned steps can obtain more image detail information in above-described embodiment after successive ignition decomposes, obtain the HFS of image, comprising raindrop and object boundary part.
In step 200, identify the edge of pictorial element in above-mentioned pending image, obtain feature contour figure.Here feature contour figure preferably carries out processing the bianry image obtained based on gray level image.In one embodiment of the invention, identify in this step that the edge of pictorial element in above-mentioned pending image obtains the process of feature contour figure, the edge detection algorithm based on gradient is adopted to carry out rim detection to image, preferably, Prewitt operator is adopted to carry out rim detection to image based on gradation of image, Prewitt operator is than being more suitable for the relatively sharp-pointed and situation that picture noise is smaller of image border gray-scale value, and its processing speed is very fast.Certain the present invention is also not limited to only adopt this kind of mode to carry out rim detection, such as, the one in the methods such as Roberts boundary operator, Sobel operator, Laplacian operator, Canny operator can also be used to carry out rim detection to pending image.
In step 300, utilize morphological image to operate and above-mentioned intramarginal image-region is processed, obtain the first intermediate image.In a preferred embodiment of the invention, above-mentioned utilize morphological image operate the process that above-mentioned intramarginal image-region processes is comprised: fill based on the Contiguous graphics region of image procossing etching operation to above-mentioned edge.Morphological scale-space in Digital Image Processing refers to and extracts for the picture content of expressing and description region shape is useful using mathematical mor-phology from image as instrument, such as border, skeleton and convex hull, also comprise for the morphologic filter of pre-service or aftertreatment, refinement and pruning etc.Wherein, for etching operation, it adds pixel can to the object bounds in image, even carries out connected region filling.The present embodiment, based on the connected region filling algorithm of image procossing etching operation, carries out holes filling to feature contour figure, obtains above-mentioned first intermediate image.
In step 400, from above-mentioned high-frequency characteristic figure, deduct above-mentioned first intermediate image, obtain rain line features figure.Adopt image arithmetic operator method to process two images in this step, improve arithmetic speed.Preferably, in one embodiment of the invention, the above-mentioned process deducting above-mentioned first intermediate image from above-mentioned high-frequency characteristic figure comprises the following steps:
First, to above-mentioned first intermediate image negate, the second intermediate image is obtained;
Secondly, be multiplied with above-mentioned high-frequency characteristic figure with above-mentioned second intermediate image, extract the common factor of above-mentioned second intermediate image and above-mentioned high-frequency characteristic figure, form above-mentioned rain line features figure.
In step 500, above-mentioned pending image and above-mentioned rain line features figure are subtracted each other, obtain the image after removing rain.Adopt image arithmetic operator method to process two images in this step, improve arithmetic speed.Preferably, in one embodiment of the invention, above-mentioned above-mentioned pending image and above-mentioned rain line features figure are subtracted each other to obtain go the process of the image after rain to comprise the following steps:
First, to above-mentioned rain line features figure negate, obtain the 3rd intermediate image;
Secondly, be multiplied with above-mentioned pending image with above-mentioned 3rd intermediate image, extract the common factor of above-mentioned 3rd intermediate image and above-mentioned pending image, formed and above-mentionedly remove the image after rain.
Based on above-described embodiment, in one embodiment of the invention, as shown in Figure 1, said method also comprises:
Step 600, above-mentioned remove rain after image on superpose above-mentioned feature contour figure, obtain repair after image.
What obtain based on above-mentioned steps 500 removes rain result figure, because contain the characteristics of image needed for some in rain line features figure, therefore can be added going rain result figure to the feature contour figure utilizing corresponding outline identification operator to carry out identifying, can access and remove rain image after reparation, improve the precision of processing result image, improve the visual effect of image.
Fig. 1 is the method flow schematic diagram of one embodiment of the invention.Although it should be understood that each step in the process flow diagram of Fig. 1 shows successively according to the instruction of arrow, these steps are not that the inevitable order according to arrow instruction performs successively.Unless had explicitly bright herein, the order that the execution of these steps is strict limits, and it can perform with other order.And, step at least partially in Fig. 1 can comprise multiple sub-step or multiple stage, these sub-steps or stage are necessarily not complete at synchronization, but can perform in the different moment, its execution sequence does not also necessarily carry out successively, but with other steps or the sub-step of other steps or the combination embodiment in stage or can exchange execution sequence.The implementation of each embodiment only for corresponding steps in illustrating is set forth above, then in the not conflicting situation of logic, each embodiment above-mentioned be can mutually combine and form new technical scheme, and this new technical scheme is still in the open scope of this embodiment.
As shown in Figure 3, present invention also offers in a kind of single image the system 800 removing raindrop impact based on said method, it comprises:
Picture breakdown module 801, for carrying out picture breakdown based on empirical mode decomposition method to pending image, extracts the HFS of described pending image, forms the high-frequency characteristic figure embodying described HFS information;
Edge detection module 802, for identifying the edge of pictorial element in described pending image, obtains feature contour figure;
Packing module 803, operates for utilizing morphological image and processes described intramarginal image-region, obtain the first intermediate image;
First computing module 804, for deducting described first intermediate image from described high-frequency characteristic figure, obtains rain line features figure; And
Second computing module 805, for described pending image and described rain line features figure being subtracted each other, obtains the image after removing rain.
Based on above-described embodiment, as shown in Figure 3, in one embodiment of the invention, said system also comprises following functions module:
Overlap-add procedure module 806, for described remove rain after image on superpose described feature contour figure, obtain repair after image.
Above-mentioned functions module 801 to 806 respectively for performing above-mentioned steps 100 to 600, its specific implementation can see above-mentioned about step 100 to the related description of 600, be not repeated at this.
In one embodiment of the invention, above-mentioned picture breakdown module 801 comprises:
Map unit, for by described pending image mapped to XOY plane, the gray-scale value of described pending image respective pixel is as Z coordinate;
Recognition unit, for being identified local maximum and the minimal value of described pending image by morphological method, obtains multiple scattered maximum point and minimum point;
Interpolating unit, for carrying out the triangulation of plane point set respectively to described multiple scattered maximum point and minimum point, then interpolation smoothing obtains maximum value envelope surface and minimal value envelope surface;
Average calculation unit, for calculating the average of described maximum value envelope surface and minimal value envelope surface;
Resolving cell, for the gray-scale value of each pixel in described pending image is deducted described average, obtains exploded view picture;
First judging unit, for judging that described exploded view similarly is no satisfied screening termination condition, if so, then exports described exploded view picture as described HFS; If not, then return and call above-mentioned recognition unit.
In one embodiment of the invention, above-mentioned picture breakdown module 801 also comprises:
Second resolving cell, for deducting described exploded view picture from described pending image, obtains when the image after pre-treatment;
Whether the second judging unit, described meet picture breakdown termination condition, if then exported as described HFS by described exploded view picture when the image after pre-treatment for judging; If not, then decompose next time, return and call above-mentioned recognition unit, until meet described picture breakdown termination condition, export the described exploded view picture repeatedly decomposing acquisition, export as described HFS.In another embodiment of the present invention, above-mentioned picture breakdown module 801 also comprises: high-frequency characteristic figure forming unit, if only comprise for above-mentioned HFS the exploded view picture that decomposable process obtains, then using this exploded view picture as above-mentioned high-frequency characteristic figure; If above-mentioned HFS comprises the exploded view picture that repeatedly decomposable process obtains respectively, then superpose the above-mentioned exploded view picture repeatedly decomposing acquisition, form above-mentioned high-frequency characteristic figure.
Each functional unit in above-mentioned picture breakdown module 801 is respectively used to perform step 101 in Fig. 2 to step 109, therefore its specific implementation can see above-mentioned about step 101 to the related description of 109, be not repeated at this.
In one embodiment of the invention, the first computing module 804 comprises:
First negate unit, for described first intermediate image negate, obtains the second intermediate image;
First multiplying unit, for being multiplied with described high-frequency characteristic figure with described second intermediate image, extracts the common factor of described second intermediate image and described high-frequency characteristic figure, forms described rain line features figure.
In one embodiment of the invention, the second computing module 805 comprises:
Second negate unit, for described rain line features figure negate, obtains the 3rd intermediate image;
Second multiplying unit, for being multiplied with described pending image with described 3rd intermediate image, extracting the common factor of described 3rd intermediate image and described pending image, removing the image after rain described in formation.
Each functional module in the system 800 of raindrop impact is removed all for performing in the above-mentioned single image shown in Fig. 1 each step of method removing raindrop impact in above-mentioned single image, its specific implementation can illustrate see the explanation of above-mentioned method step, is not repeated at this.
Through the above description of the embodiments, those skilled in the art can be well understood to the mode that above-described embodiment method can add required general hardware platform by software and realize, hardware can certainly be passed through, but in a lot of situation, the former is better embodiment.Based on such understanding, technical scheme of the present invention can embody with the form of software product the part that prior art contributes in essence in other words, this computer software product is carried on a non-volatile computer readable storage medium (as ROM, magnetic disc, CD, server storage) in, comprising some instructions in order to make a station terminal equipment (can be mobile phone, computing machine, server, or the network equipment etc.) perform system architecture described in each embodiment of the present invention and method.
In sum, the present invention proposes a kind of new image in this article and goes rain method, it is by the picture breakdown technology based on empirical mode decomposition, obtain the HFS of image, re-use image border recognizer, both subtract each other at result, thus finally obtain the pixel of being rung by rain shadow, because the pixel intensity of being rung by rain shadow is higher, deduct it finally by from former figure, obtain rain image.The present invention effectively can improve the visual effect of ringing image by rain shadow, can process coloured image, and improve arithmetic speed.Further, use experience Mode Decomposition of the present invention carries out picture breakdown, and based on image arithmetic operator to getting process of occuring simultaneously between image, has the performance being better than other algorithms, and the time significantly reduced needed for rain, well below the time of conventional process.The single image that present invention overcomes based on sparse coding goes the shortcoming that only can process gray level image of rain algorithm, can process, and can obtain good effect to coloured image.Such as, the pending image shown in Fig. 4, obtains design sketch as shown in Figure 5 after the process of above-mentioned steps 100 to the step 600 based on two-dimensional empirical mode decomposition method and Prewitt limb recognition algorithm, and the time of process piece image is 25.473 seconds.
The above embodiment only have expressed several embodiment of the present invention, and it describes comparatively concrete and detailed, but therefore can not be interpreted as the restriction to the scope of the claims of the present invention.It should be pointed out that for the person of ordinary skill of the art, without departing from the inventive concept of the premise, can also make some distortion and improvement, these all belong to protection scope of the present invention.Therefore, the protection domain of patent of the present invention should be as the criterion with claims.
Claims (10)
1. remove a method for raindrop impact in single image, it comprises:
Based on empirical mode decomposition method, picture breakdown is carried out to pending image, extract the HFS of described pending image, form the high-frequency characteristic figure embodying described HFS information;
Identify the edge of pictorial element in described pending image, obtain feature contour figure;
Utilize morphological image to operate to process described intramarginal image-region, obtain the first intermediate image;
From described high-frequency characteristic figure, deduct described first intermediate image, obtain rain line features figure;
Described pending image and described rain line features figure are subtracted each other, obtains the image after removing rain.
2. remove the method for raindrop impact in single image according to claim 1, it is characterized in that, described method also comprises:
Described remove rain after image on superpose described feature contour figure, obtain repair after image.
3. the process of the HFS remove the method for raindrop impact in single image according to claim 1, it is characterized in that, describedly carry out picture breakdown based on empirical mode decomposition method to pending image, extracting described pending image comprises the following steps:
By described pending image mapped to XOY plane, the gray-scale value of described pending image respective pixel is as Z coordinate;
Identified local maximum and the minimal value of described pending image by morphological method, obtain multiple scattered maximum point and minimum point;
Described multiple scattered maximum point and minimum point are carried out respectively to the triangulation of plane point set, then interpolation smoothing obtains maximum value envelope surface and minimal value envelope surface;
Calculate the average of described maximum value envelope surface and minimal value envelope surface;
The gray-scale value of each pixel in described pending image is deducted described average, obtains exploded view picture;
Judge that described exploded view similarly is no satisfied screening termination condition, if so, then described exploded view picture is exported as described HFS; If not, then return described in execution and identify the local maximum of described pending image and minimizing step by morphological method.
4. remove the method for raindrop impact in single image according to claim 3, it is characterized in that, described described exploded view picture also to be comprised before the step that described HFS exports:
From described pending image, deduct described exploded view picture, obtain when the image after pre-treatment;
When whether the image after pre-treatment meets picture breakdown termination condition, if then exported as described HFS by described exploded view picture described in judging;
If not, then decompose next time, return to perform and identify the local maximum of described pending image and minimizing step by morphological method, until meet described picture breakdown termination condition, export and repeatedly decompose the described exploded view picture obtained respectively, export as described HFS.
5. in single image according to claim 1, remove the method for raindrop impact, it is characterized in that, in the described pending image of described identification, the edge of pictorial element obtains in the process of feature contour figure, adopts Privett operator to carry out rim detection to image based on gradation of image.
6. in single image according to claim 1, remove the method for raindrop impact, it is characterized in that, described utilize morphological image operate the process that described intramarginal image-region processes is comprised: fill based on the Contiguous graphics region of image procossing etching operation to described edge.
7. remove the method for raindrop impact in single image according to claim 1, it is characterized in that, describedly from described high-frequency characteristic figure, deduct the process that described first intermediate image obtains rain line features figure comprise:
To described first intermediate image negate, obtain the second intermediate image;
Be multiplied with described high-frequency characteristic figure with described second intermediate image, extract the common factor of described second intermediate image and described high-frequency characteristic figure, form described rain line features figure.
8. remove in single image according to claim 1 raindrop impact method, it is characterized in that, described described pending image and described rain line features figure are subtracted each other to obtain go the process of the image after rain to comprise:
To described rain line features figure negate, obtain the 3rd intermediate image;
Be multiplied with described pending image with described 3rd intermediate image, extract the common factor of described 3rd intermediate image and described pending image, described in formation, remove the image after rain.
9. remove a system for raindrop impact in single image, it is characterized in that, described system comprises:
Picture breakdown module, for carrying out picture breakdown based on empirical mode decomposition method to pending image, extracts the HFS of described pending image, forms the high-frequency characteristic figure embodying described HFS information;
Edge detection module, for identifying the edge of pictorial element in described pending image, obtains feature contour figure;
Packing module, operates for utilizing morphological image and processes described intramarginal image-region, obtain the first intermediate image;
First computing module, for deducting described first intermediate image from described high-frequency characteristic figure, obtains rain line features figure; And
Second computing module, for described pending image and described rain line features figure being subtracted each other, obtains the image after removing rain.
10. remove the system of raindrop impact in single image according to claim 9, it is characterized in that, described system also comprises:
Overlap-add procedure module, for described remove rain after image on superpose described feature contour figure, obtain repair after image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410856519.XA CN104537622B (en) | 2014-12-31 | 2014-12-31 | The method and system that raindrop influence is removed in single image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410856519.XA CN104537622B (en) | 2014-12-31 | 2014-12-31 | The method and system that raindrop influence is removed in single image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104537622A true CN104537622A (en) | 2015-04-22 |
| CN104537622B CN104537622B (en) | 2018-02-13 |
Family
ID=52853141
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410856519.XA Active CN104537622B (en) | 2014-12-31 | 2014-12-31 | The method and system that raindrop influence is removed in single image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104537622B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104966275A (en) * | 2015-06-12 | 2015-10-07 | 中国科学院深圳先进技术研究院 | Method and system for removing raindrop influence from single image based on rain frequency characteristics |
| CN104978720A (en) * | 2015-07-01 | 2015-10-14 | 深圳先进技术研究院 | Video image raindrop removal method and apparatus |
| CN105139358A (en) * | 2015-08-28 | 2015-12-09 | 中国科学院深圳先进技术研究院 | Video raindrop removing method and system based on combination of morphology and fuzzy C clustering |
| CN105139344A (en) * | 2015-06-12 | 2015-12-09 | 中国科学院深圳先进技术研究院 | Method and system for removing influence of raindrops to single image based on frequency domain and phase consistency |
| CN109360155A (en) * | 2018-08-17 | 2019-02-19 | 上海交通大学 | Single-frame images rain removing method based on multi-scale feature fusion |
| CN112085680A (en) * | 2020-09-09 | 2020-12-15 | 腾讯科技（深圳）有限公司 | Image processing method and device, electronic equipment and storage medium |
| CN113223039A (en) * | 2020-01-21 | 2021-08-06 | 海信集团有限公司 | Display device, clothing image extraction method, and storage medium |
| CN116310598A (en) * | 2023-05-16 | 2023-06-23 | 常州海图信息科技股份有限公司 | Obstacle detection method and device for severe weather |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20090089152A (en) * | 2008-02-18 | 2009-08-21 | 성균관대학교산학협력단 | Caption extraction system and control method thereof |
| CN102186089A (en) * | 2011-04-19 | 2011-09-14 | 南京航空航天大学 | Simple-component video image rain field removing method |
| CN103226813A (en) * | 2013-03-29 | 2013-07-31 | 南通大学 | Processing method for improving video image quality in rainy days |
| CN103337061A (en) * | 2013-07-18 | 2013-10-02 | 厦门大学 | Rain and snow removing method for image based on multiple guided filtering |
| CN103700070A (en) * | 2013-12-12 | 2014-04-02 | 中国科学院深圳先进技术研究院 | Video raindrop-removing algorithm based on rain-tendency scale |
| CN104112259A (en) * | 2014-06-19 | 2014-10-22 | 中国科学院深圳先进技术研究院 | Rain removing method and system for single image |
| CN104112255A (en) * | 2014-06-19 | 2014-10-22 | 中国科学院深圳先进技术研究院 | Rain removing method and system for single image |
-
2014
- 2014-12-31 CN CN201410856519.XA patent/CN104537622B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20090089152A (en) * | 2008-02-18 | 2009-08-21 | 성균관대학교산학협력단 | Caption extraction system and control method thereof |
| CN102186089A (en) * | 2011-04-19 | 2011-09-14 | 南京航空航天大学 | Simple-component video image rain field removing method |
| CN103226813A (en) * | 2013-03-29 | 2013-07-31 | 南通大学 | Processing method for improving video image quality in rainy days |
| CN103337061A (en) * | 2013-07-18 | 2013-10-02 | 厦门大学 | Rain and snow removing method for image based on multiple guided filtering |
| CN103700070A (en) * | 2013-12-12 | 2014-04-02 | 中国科学院深圳先进技术研究院 | Video raindrop-removing algorithm based on rain-tendency scale |
| CN104112259A (en) * | 2014-06-19 | 2014-10-22 | 中国科学院深圳先进技术研究院 | Rain removing method and system for single image |
| CN104112255A (en) * | 2014-06-19 | 2014-10-22 | 中国科学院深圳先进技术研究院 | Rain removing method and system for single image |
Non-Patent Citations (6)
| Title |
|---|
| QINGSONG ZHU 等: "a novel rain detection and removal approach using Guided Filtering and Formation Modeling", 《PROCEEDING OF THE IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND BIOMIMETICS(ROBIO),SHENZHEN,CHINA》 * |
| XIANHUI ZHENG 等: "Single-Image-Based Rain and Snow Removal Using Multi-guided Filter", 《20TH INTERNATIONAL CONFERENCE ON NEURAL INFORMATION PROCESSING (ICONIP)》 * |
| 冯瑞利: "雨雾天气图像的清晰化技术研究", 《中国优秀硕士学位论文全文数据库_信息科技辑》 * |
| 成蕾: "视频图像去雨雾算法的研究", 《中国优秀硕士学位论文全文数据库_信息科技辑》 * |
| 秦绪佳 等: "一种边缘保持的医学图像去噪方法", 《计算机科学》 * |
| 赵小川: "《MATLAB图像处理:能力提高与应用案例》", 31 January 2014 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104966275A (en) * | 2015-06-12 | 2015-10-07 | 中国科学院深圳先进技术研究院 | Method and system for removing raindrop influence from single image based on rain frequency characteristics |
| CN105139344A (en) * | 2015-06-12 | 2015-12-09 | 中国科学院深圳先进技术研究院 | Method and system for removing influence of raindrops to single image based on frequency domain and phase consistency |
| CN104966275B (en) * | 2015-06-12 | 2018-01-05 | 中国科学院深圳先进技术研究院 | Single image based on rain frequency characteristic removes the method and system that raindrop influence |
| CN104978720A (en) * | 2015-07-01 | 2015-10-14 | 深圳先进技术研究院 | Video image raindrop removal method and apparatus |
| CN105139358A (en) * | 2015-08-28 | 2015-12-09 | 中国科学院深圳先进技术研究院 | Video raindrop removing method and system based on combination of morphology and fuzzy C clustering |
| CN109360155B (en) * | 2018-08-17 | 2020-10-13 | 上海交通大学 | Single-frame image rain removing method based on multi-scale feature fusion |
| CN109360155A (en) * | 2018-08-17 | 2019-02-19 | 上海交通大学 | Single-frame images rain removing method based on multi-scale feature fusion |
| CN113223039A (en) * | 2020-01-21 | 2021-08-06 | 海信集团有限公司 | Display device, clothing image extraction method, and storage medium |
| CN113223039B (en) * | 2020-01-21 | 2023-04-07 | 海信集团有限公司 | Display device, clothing image extraction method, and storage medium |
| CN112085680A (en) * | 2020-09-09 | 2020-12-15 | 腾讯科技（深圳）有限公司 | Image processing method and device, electronic equipment and storage medium |
| CN112085680B (en) * | 2020-09-09 | 2023-12-12 | 腾讯科技（深圳）有限公司 | Image processing method and device, electronic equipment and storage medium |
| CN116310598A (en) * | 2023-05-16 | 2023-06-23 | 常州海图信息科技股份有限公司 | Obstacle detection method and device for severe weather |
| CN116310598B (en) * | 2023-05-16 | 2023-08-22 | 常州海图信息科技股份有限公司 | Obstacle detection method and device for severe weather |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104537622B (en) | 2018-02-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104537622A (en) | Method and system for removing raindrop influence in single image | |
| Pei et al. | Removing rain and snow in a single image using saturation and visibility features | |
| Liu et al. | Single image dehazing via large sky region segmentation and multiscale opening dark channel model | |
| CN104599256B (en) | The method and system of removal image rain line based on single image | |
| Fei et al. | Visual tracking based on improved foreground detection and perceptual hashing | |
| CN103578084A (en) | Color image enhancement method based on bright channel filtering | |
| CN109558806A (en) | The detection method and system of high score Remote Sensing Imagery Change | |
| CN111861925A (en) | Image rain removing method based on attention mechanism and gate control circulation unit | |
| CN103745468A (en) | Significant object detecting method based on graph structure and boundary apriority | |
| CN105139344B (en) | The method and system influenced based on frequency domain and the single image of phase equalization removal raindrop | |
| WO2021118463A1 (en) | Defect detection in image space | |
| Lian et al. | A novel method on moving-objects detection based on background subtraction and three frames differencing | |
| CN103578083A (en) | Single image defogging method based on joint mean shift | |
| CN109064419A (en) | A kind of removing rain based on single image method based on WLS filtering and multiple dimensioned sparse expression | |
| Acharjya et al. | Watershed segmentation based on distance transform and edge detection techniques | |
| Koo et al. | Nighttime haze removal with glow decomposition using GAN | |
| Wang et al. | Haze removal algorithm based on single-images with chromatic properties | |
| Fengping et al. | Road extraction using modified dark channel prior and neighborhood FCM in foggy aerial images | |
| Harianto et al. | Data augmentation and faster rcnn improve vehicle detection and recognition | |
| Chen et al. | Visual depth guided image rain streaks removal via sparse coding | |
| Pal et al. | Visibility enhancement of images degraded by hazy weather conditions using modified non-local approach | |
| Niu et al. | Image deblurring of video surveillance system in rainy environment | |
| CN109241865B (en) | Vehicle detection segmentation algorithm under weak contrast traffic scene | |
| Chowdhury et al. | A background subtraction method using color information in the frame averaging process | |
| Taha et al. | Moving shadow removal for multi-objects tracking in outdoor environments |
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 |