CN105096316A - Method and device for determining image blurriness - Google Patents
Method and device for determining image blurriness Download PDFInfo
- Publication number
- CN105096316A CN105096316A CN201510376667.6A CN201510376667A CN105096316A CN 105096316 A CN105096316 A CN 105096316A CN 201510376667 A CN201510376667 A CN 201510376667A CN 105096316 A CN105096316 A CN 105096316A
- Authority
- CN
- China
- Prior art keywords
- image block
- fuzzy
- sparse
- parameter
- described 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30168—Image quality inspection
Landscapes
- Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The invention discloses a method and a device for determining image blurriness, which is used for solving the problem of obtaining a reliable blurriness evaluation value of an image area with unremarkable blurriness difference with the clear area of the image is hard. The method for determining image blurriness comprises steps of performing sparse coding on image blocks of the image according to a sparse dictionary so as to obtain a sparse vector showing the image block, determining the sparse characteristics of the image block according to the sparse vector of the image block, and determining the blurriness parameter of the image block according to the sparse characteristics, wherein the blurriness parameter shows the blurriness of the image block.
Description
Technical field
The present invention relates to image processing field, particularly a kind of method and device determining image blurring degree.
Background technology
The zones of different of image has different readabilitys, and namely the zones of different of image has different fog-levels.Determine the fog-level of the zones of different of image, have great help for image procossing such as such as Postprocessing technique, image enhaucament, Iamge Segmentation, image recognitions.
In prior art, utilize the otherness had of the fuzzy region in image and clear area to determine the fog-level of image-region, such as, utilize the Gradient Features of image, or the spectrum signature of image determines the fog-level of image-region.
But, the fog-level of image-region is determined based on the clear area of image and the otherness of fuzzy region, need when the size of this difference meets some requirements, just can obtain the valuation of reliable image-region fog-level, when this otherness is too small and when cannot meet above-mentioned condition, adopt above-mentioned prior art cannot obtain the valuation of reliable image-region fog-level.
Summary of the invention
The embodiment of the present invention provides a kind of method and the device of determining image blurring degree, is difficult to obtain the problem with the reliable fog-level valuation of the inapparent image-region of fog-level difference of the clear area of image for solving.
First aspect, the embodiment of the present invention provides a kind of method determining image blurring degree, comprising:
According to sparse dictionary, sparse coding is carried out to the image block of described image, thus obtain the sparse vector for representing described image block;
According to the sparse vector of described image block, determine the sparse features of described image block;
According to described sparse features, determine the fuzzy parameter of described image block, described fuzzy parameter represents the fog-level of described image block.
In conjunction with first aspect, in the first possible implementation of first aspect, described according to described sparse features, determine the fuzzy parameter of described image block, comprising:
According to the corresponding relation between described sparse features and described fuzzy parameter, and described sparse features, determine described fuzzy parameter.
In conjunction with the first possible implementation of first aspect or first aspect, in the implementation that the second of first aspect is possible, described according to described sparse features, determine the fuzzy parameter of described image block, comprising:
Described fuzzy parameter is determined by solving following equation:
Wherein, f is described sparse features, and σ is described fuzzy parameter, and a, b, c, d, e are preset value.
In conjunction with the first possible implementation of first aspect, first aspect to any one in the possible implementation of the second of first aspect, in the third possible implementation of first aspect, the described sparse vector according to described image block, determine the sparse features of described image block, comprising:
Calculate zero norm of the sparse vector of described image block, using described zero norm as described sparse features.
In conjunction with the first possible implementation of first aspect, first aspect to any one in the third possible implementation of first aspect, in the 4th kind of possible implementation of first aspect, described determine the fuzzy parameter of described image block after, described method also comprises:
Judge described image block whether in focal zone again;
When described image block is not in described focal zone again, Fuzzy Processing corresponding to described fuzzy parameter is carried out to described image block.
In conjunction with the 4th kind of possible implementation of first aspect, in the 5th kind of possible implementation of first aspect, describedly judge described image block whether in focal zone again, comprising:
Obtain the minimum depth value of described image; Using the depth value of described fuzzy parameter as described image block, judge whether the difference that the depth value of described image block deducts described minimum depth value is not less than setting threshold value; If the difference that the depth value of described image block deducts described minimum depth value is not less than described setting threshold value, then determine described image block not in described focal zone again.
In conjunction with the first possible implementation of first aspect, first aspect to any one in the third possible implementation of first aspect, in the 6th kind of possible implementation of first aspect, described determine the fuzzy parameter of described image block after, described method also comprises:
Using the fuzzy core parameter of described fuzzy parameter as described image block, deblurring process corresponding to described fuzzy core parameter is carried out to described image block.
In conjunction with the first possible implementation of first aspect, first aspect to any one in the third possible implementation of first aspect, in the 7th kind of possible implementation of first aspect, described determine the fuzzy parameter of described image block after, described method also comprises:
According to described fuzzy parameter, synthesis corresponds to the fuzzy graph of described image, and the pixel value corresponding to the region of described image block in described fuzzy graph is described fuzzy parameter;
According to described fuzzy graph, determine the depth value of described image.
Second aspect, the embodiment of the present invention provides a kind of device determining image blurring degree, comprising:
Sparse coding module, for carrying out sparse coding according to sparse dictionary to the image block of described image, thus obtains the sparse vector for representing described image block;
First determination module, for the sparse vector of the described image block according to described sparse coding module acquisition, determines the sparse features of described image block;
Second determination module, for the described sparse features obtained according to described first determination module, determine the fuzzy parameter of described image block, described fuzzy parameter represents the fog-level of described image block.
In conjunction with second aspect, in the first possible implementation of second aspect, described second determination module is used for: according to the corresponding relation between described sparse features and described fuzzy parameter, and the described sparse features that described first determination module obtains, and determines described fuzzy parameter.
In conjunction with the first possible implementation of second aspect or second aspect, in the implementation that the second of second aspect is possible, described second determination module is used for: determine described fuzzy parameter by solving following equation:
Wherein, f is described sparse features, and σ is described fuzzy parameter, and a, b, c, d, e are preset value.
In conjunction with the first possible implementation of second aspect, second aspect to any one in the possible implementation of the second of second aspect, in the third possible implementation of second aspect, described first determination module is used for: zero norm calculating the sparse vector of described image block, using described zero norm as described sparse features.
In conjunction with the first possible implementation of first aspect, first aspect to any one in the third possible implementation of first aspect, in the 4th kind of possible implementation of first aspect, described device also comprises:
Fuzzy Processing module, for judging described image block whether in focal zone again; And when described image block is not in described focal zone again, Fuzzy Processing corresponding to described fuzzy parameter is carried out to described image block.
In conjunction with the 4th kind of possible implementation of first aspect, in the 5th kind of possible implementation of second aspect, described Fuzzy Processing module is used for: the minimum depth value obtaining described image; Using the depth value of described fuzzy parameter as described image block, judge whether the difference that the depth value of described image block deducts described minimum depth value is not less than setting threshold value; If the difference that the depth value of described image block deducts described minimum depth value is not less than described setting threshold value, then determines described image block not in described focal zone again, Fuzzy Processing corresponding to described fuzzy parameter is carried out to described image block.
In conjunction with the first possible implementation of second aspect, second aspect to any one in the third possible implementation of second aspect, in the 6th kind of possible implementation of second aspect, described device also comprises:
Deblurring module, for using the fuzzy core parameter of described fuzzy parameter as described image block, carries out deblurring process corresponding to described fuzzy parameter to described image block.
In conjunction with the first possible implementation of second aspect, second aspect to any one in the third possible implementation of second aspect, in the 7th kind of possible implementation of second aspect, described device also comprises:
3rd determination module, for according to described fuzzy parameter, synthesize the fuzzy graph corresponding to described image, the pixel value corresponding to the region of described image block in described fuzzy graph is described fuzzy parameter; And according to described fuzzy graph, determine the depth value of described image.
The one or more technical schemes provided in the embodiment of the present invention, at least have following technique effect or advantage:
In the embodiment of the present invention, by utilizing sparse dictionary to encode to image block, obtain the sparse vector of image block, and then obtain the sparse features of image, and according to the fuzzy parameter of sparse features determination image block, and then know the fog-level of image block.The significant difference of the fog-level of the clear area of image block and image is not relied on due to said method, can in the inapparent situation of fog-level difference of the clear area of image block and image, determine the fuzzy parameter of image block, reliable valuation is carried out to the fog-level of image block, implementation is simple, and result accurately and reliably.
Term " first ", " second ", " the 3rd " " 4th " etc. (if existence) in instructions of the present invention and claims and above-mentioned accompanying drawing are for distinguishing similar object, and need not be used for describing specific order or precedence.Should be appreciated that the data used like this can be exchanged in the appropriate case, so as embodiments of the invention described herein such as can with except here diagram or describe those except order implement.In addition, term " comprises " and " having " and their any distortion, intention is to cover not exclusive comprising, such as, contain those steps or unit that the process of series of steps or unit, method, system, product or equipment is not necessarily limited to clearly list, but can comprise clearly do not list or for intrinsic other step of these processes, method, product or equipment or unit.
For the ease of understanding the technical scheme that the embodiment of the present invention provides, first introduce the sparse coding of image.
The sparse coding of image refers to that image block can be represented by the linear combination of considerably less one group of atom (Atom) image block completely or approx, and namely all atomic lens blocks form an excessively complete dictionary, the number forming the atom of this dictionary is greater than the dimension of each atom, and this is crossed complete dictionary and is sparse dictionary.
Note N dimensional vector y ∈ R
nfor the vector representation of image block, D is the set of K N dimensional vector, is designated as: D={g
i}
i ∈ 1,2 ..., k, N dimensional vector g
i∈ R
nbe called atom, set D is called dictionary.As K > N, atom linear correlation, dictionary D redundancy, is called sparse dictionary.
N dimensional vector y ∈ R
nthe linear combination of sparse dictionary D Atom can be expressed as:
Or vectorial y can close approximation be:
Wherein, C=(c
1, c
2..., c
k) being the sparse vector of y in sparse dictionary D, R is for approaching remnants.
Because the atom number of sparse dictionary is greater than atom dimension, if to rarefaction representation without any restriction, so the sparse vector of image block in sparse dictionary is not unique.In actual conditions, in order to obtain unique sparse vector, usually need to arrange certain restrictive condition.
Such as, restrictive condition can be: min||C||
1, s.t.||y-DC||
2≤ ε, wherein, the computing of min [] for minimizing, s.t. is the abbreviation of subjectto, and the implication of " minA, s.t.B " is: the minimum value asking A under constraint condition B, || ||
prepresent p-norm, when p gets 1, || C||
1=| c
1|+| c
2|+...+| c
k|; When p gets 2,
under above-mentioned constraint condition, the unique sparse vector C of vectorial y in sparse dictionary D can be obtained.
With reference to Fig. 1, the schematic flow sheet of the method for the image blurring degree of determination provided for the embodiment of the present invention, this flow process comprises the steps:
Step 101: sparse coding is carried out to the image block of image according to sparse dictionary, thus obtain the sparse vector for representing image block;
Step 102: according to the sparse vector of image block, determines the sparse features of image block;
Step 103: according to sparse features, determines the fuzzy parameter of image block, and this fuzzy parameter represents the fog-level of image block.
Concrete, in step 101, processed image can be the image that the image acquisition units of the electronic equipment of the method performing the embodiment of the present invention gathers, and also can be this electronic equipment from the image of other electronic equipments or network reception.
The image block of image refers to: be minimum unit with pixel, the block of the m*n marked off from image.With reference to Fig. 2, for the image of 4*5 being divided into the example of the image block a ~ f of 6 3*3, when partitioned image block, make two adjacent image blocks except a line (or row) is except pixel is not overlapping, other pixels are overlapping, and in figure, each arabic numeral represent a pixel, such as, in block a and block b, pixel 2,3,7,8,12,13 is overlapping.In actual conditions, two adjacent image blocks can be completely not overlapping yet, or carry out overlap with lower degree of overlapping.Should be understood that the image block division methods shown in Fig. 2 is only the exemplary illustration of the embodiment of the present invention, other methods that image can be divided into image block all fall within the scope of protection of the present invention.
Image block can be expressed as vector, as the image block a marked off in Fig. 2, can be expressed as vector (h
1, h
2, h
3, h
6, h
7, h
8, h
11, h
12, h
13), wherein h
tbe the gray-scale value of t pixel, in actual conditions, except gray-scale value, also can represent the vector of image block with the pixel value of color space, be not described in detail in this.
As aforementioned to the introduction of sparse coding, can encode with the vector of sparse dictionary to image block, obtain sparse vector.In the embodiment of the present invention, sparse dictionary can be kept at electronic equipment this locality, and when performing step 101, electronic equipment stores from this locality and reads sparse dictionary.Sparse dictionary also can be kept at and can carry out on the miscellaneous equipment of data interaction with electronic equipment, and when performing step 101, electronic equipment can send request to the equipment preserving this sparse dictionary, obtains this sparse dictionary.
In the embodiment of the present invention, sparse dictionary comprises the information of blurred picture, and this sparse dictionary can be generated by study, also can be generated by mathematical model, manually can also generate for developer, the generating mode of sparse dictionary please refer to prior art, and embodiment of the present invention example will not limit.
In step 101, when carrying out sparse coding to image block, need certain sparse restrictive condition, such as, sparse restrictive condition can be: min||C||
1, s.t.||y-DC||
2≤ ε, wherein, ε is the threshold value of setting, for control image block and sparse dictionary atom linear combination between approximation ratio.
Again such as, note μ is the minimum value of dictionary D Atom linear correlation, and sparse restrictive condition can be:
wherein, || C||
0represent the number of nonzero element in vectorial C, be called the sparse features of vectorial C.During concrete enforcement, sparse restrictive condition can also have other form, can with reference to various feasible ways to restrain of the prior art, and the embodiment of the present invention will not be illustrated one by one.
The sparse vector solving image block under sparse restrictive condition is the expansion of the representation such as Discrete Fourier Transformation of Digital Images, wavelet transformation, and it solves mode and comprises: matching pursuit algorithm, base tracing algorithm, iteration focusing algorithm, greedy algorithm etc.About the mode of sparse vector solving image block under sparse restrictive condition, please refer to prior art, the embodiment of the present invention will not describe in detail.
In step 102, the sparse features of image block can be zero norm of the sparse vector of image block, and namely the number of nonzero element in sparse vector, is designated as: f=||C||
0.According to the sparse vector of the image block obtained in step 101, the sparse features of image block can be determined.
Then, step 103 is performed, according to the fuzzy parameter of the coefficient characteristics determination image block of image block.
Concrete, present inventor is after carrying out great many of experiments, find that the sparse features of image block and the fuzzy parameter of this image block have correlationship, therefore, corresponding relation between the fuzzy parameter can being determined sparse features and this image block by a large amount of experimental datas, this corresponding relation can be functional relation, also can be the corresponding lists of sparse features and fuzzy parameter, after determining the sparse features of image block, by inquiring about this corresponding lists, the fuzzy parameter of image block can be determined.
In the embodiment of the present invention, the corresponding relation between sparse features and fuzzy parameter can be preserved in electronic equipment, also can be that the corresponding relation between sparse features with fuzzy parameter is preserved on the electronic equipment miscellaneous equipment that can communicate with it, electronic equipment, by sending request to the equipment preserving corresponding relation, obtains this corresponding relation.
When performing step 103, according to the corresponding relation between the fuzzy parameter of the sparse features of the image block determined, the sparse features prestored and image block, the fuzzy parameter of the fog-level for token image block can be determined.Wherein, fuzzy parameter is specifically as follows Gaussian Blur parameter, fuzzy entropy parameter, etc.
In technique scheme, by utilizing sparse dictionary to encode to image block, obtain the sparse vector of image block, and then obtain the sparse features of image, and according to the fuzzy parameter of sparse features determination image block, and then know the fog-level of image block.The significant difference of the fog-level of the clear area of image block and image is not relied on due to said method, can in the inapparent situation of fog-level difference of the clear area of image block and image, determine the fuzzy parameter of image block, reliable valuation is carried out to the fog-level of image block, implementation is simple, and result accurately and reliably
Optionally, in the embodiment of the present invention, step 103: according to sparse features, determines the fuzzy parameter of image block, when specifically implementing, comprises the steps: according to the corresponding relation between sparse features and fuzzy parameter, and sparse features, determines fuzzy parameter.
Optionally, in the embodiment of the present invention, when performing step 103, calculate fuzzy parameter especially by solving following equation:
Wherein, f is sparse features, and σ is fuzzy parameter, and a, b, c, d, e are preset value.
Wherein, the one of a, b, c, d, e may value be 39.49, Isosorbide-5-Nitrae .535 ,-3.538,38.53.
The fuzzy parameter of image block can be determined according to the sparse features of image block, to carry out image procossing better by above-mentioned relation formula.
Optionally, in the embodiment of the present invention, in step 102, according to sparse vector, determine the sparse features of image block, when specifically implementing, comprise the steps: zero norm of compute sparse vector, using this zero norm as sparse features.
Optionally, in the embodiment of the present invention, the sparse dictionary in step 101 can carry out study according to the fuzzy graph image set that sparse vector is known as learning sample and obtain.
Concrete, fuzzy graph image set is made up of the blurred picture sample that sparse vector is known, and blurred picture sample exists the image of blurred picture block and picture rich in detail block, and wherein, the fog-level difference of picture rich in detail block and blurred picture interblock can be very little.The mode obtaining sparse dictionary according to blurred picture sample learning includes but not limited to: principal component analysis (PCA) is (English: PrincipalComponentAnalysis; PCA), optimal direction method (MethodofOptimalDirections be called for short:; Be called for short: MOD), singular value decomposition method is (English: Singularvaluedecomposition; Be called for short: SVD), etc.Specifically please refer to prior art, the embodiment of the present invention will not describe in detail.In actual conditions, along with the blurred picture sample of study is on the increase, sparse dictionary is constantly updated, and can carry out preferably sparse coding to blurred picture, and the sparse coding of the blurred picture sample learnt before under sparse dictionary may upgrade along with the renewal of number of words dictionary.
Accompanying drawing explanation
In order to be illustrated more clearly in the technical scheme in the embodiment of the present invention, below the accompanying drawing used required in describing embodiment is briefly introduced, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skill in the art, under the prerequisite not paying creative work, other accompanying drawing can also be obtained according to these accompanying drawings.
Fig. 1 is a kind of schematic flow sheet determining the method for image blurring degree in the embodiment of the present invention;
Fig. 2 is the schematic diagram of the dividing mode of image block in the embodiment of the present invention;
Fig. 3 utilizes fuzzy parameter to carry out the schematic flow sheet of Fuzzy Processing to image in the embodiment of the present invention;
Fig. 4 utilizes fuzzy parameter to carry out the schematic flow sheet of deblurring process to image in the embodiment of the present invention;
Fig. 5 is the schematic flow sheet utilizing fuzzy parameter determination picture depth in the embodiment of the present invention;
Fig. 6 is the structural schematic block diagram of device 200 in the embodiment of the present invention;
Fig. 7 is the structural schematic block diagram of electronic equipment 300 in the embodiment of the present invention.
Embodiment
Below by accompanying drawing and specific embodiment, technical solution of the present invention is described in detail, the specific features being to be understood that in the embodiment of the present invention and embodiment is the detailed description to technical solution of the present invention, instead of the restriction to technical solution of the present invention, when not conflicting, the technical characteristic in the embodiment of the present invention and embodiment can combine mutually.
During concrete enforcement, can pre-determining and preserve sparse dictionary, when performing step 101, reading this sparse dictionary from storage space.
Optionally, in the embodiment of the present invention, in a step 101, the dimension of the image block of image should adapt with the dimension of sparse dictionary, to ensure the correctness of rarefaction representation.Such as, when the atom of sparse dictionary is 9 dimensional vector, image block can be the image block of 3*3.
Optionally, in the embodiment of the present invention, after step 103, step can also be comprised: the fuzzy parameter according to the described each image block determined carries out the image procossing corresponding with the fuzzy parameter of image block to image block.
Concrete, fuzzy parameter is the important indicator of image, after the fuzzy parameter of each image block determining image, can know the fog-level of each image block of image.Using the fuzzy parameter of image block as auxiliary parameter when image being carried out to image procossing, the image after processing can be made to seem more natural, more level and smooth, have more whole harmonious.
Such as, when carrying out super-resolution reconstruction to image, insert more pixel in the image block that fog-level is larger, and in the image block that fog-level is less, insert less pixel, like this, in the image after process, the sharpness of zones of different presents consistance, and image is more harmonious.
Again such as, Fuzzy Processing is carried out in some regions (namely in image, virtualization process) time, the less virtualization of intensity is carried out for the image block that fog-level is less, and the image block larger to fog-level carries out the larger image block of intensity, so make virtualization region seem more natural, more level and smooth, have more whole harmonious.
Optionally, in the embodiment of the present invention, with reference to Fig. 3, after step 103, also comprise:
Step 104: judge image block whether in focal zone again;
Step 105: when image block is not in focal zone again, Fuzzy Processing corresponding to fuzzy parameter is carried out to image block.
Concrete, again focal zone refers to: to the first area of image do not process (or, carry out image enhancement processing), and Fuzzy Processing is carried out to region remaining in image, and after above-mentioned process, fogging of the region beyond first area, by contrast, becoming in the picture of first area is more outstanding, is equivalent to, using the focal zone of first area as image, be focal zone again.
In the embodiment of the present invention, focal zone can be determined by the manual operation of user again, and such as, the arbitrary region in the image of user's point touching screen display, as focal zone again; Also can carry out input operation for user by the peripheral hardware such as mouse, keyboard input equipment, determine that a region is as focal zone again in the picture.
For image block arbitrary in image, first judge it whether within focal zone again, if not within focal zone again, then carry out Fuzzy Processing to it.Wherein, when carrying out Fuzzy Processing to it, the fuzzy parameter (that is, the fog-level of image block) according to this image block determines the intensity of it being carried out to Fuzzy Processing.
Such as, when being treated fuzzy region by wave filter and carrying out filtered blurry process, for the image block of fuzzy parameter comparatively large (corresponding fog-level is higher), filtering strength can be less, and for the less image block of fuzzy parameter, the filtering strength of wave filter can be corresponding comparatively large, so make fuzzy after image more natural, more harmonious.
Optionally, when carrying out Fuzzy Processing to the region outside focal zone again, the fuzzy parameter of arbitrary image block in the region outside focal zone is made again to be all greater than again the fuzzy parameter of focal zone, with the characteristics of image making image emphasis present again focal zone.
Optionally, in the embodiment of the present invention, step 104: judge image block whether in focal zone again, also comprise following embodiment:
Obtain the minimum depth value of image; Using the depth value of fuzzy parameter as image block, judge whether the difference that the depth value of image block deducts minimum depth value is not less than setting threshold value; If the difference that the depth value of image block deducts minimum depth value is not less than setting threshold value, then determine image block not in focal zone again.
Concrete, present inventor finds that the fog-level of image block and the depth of field of image block are proportionate, that is, image block depth value is larger, and the fuzzy parameter of image block is larger.Therefore, the depth value of the fuzzy parameter token image block of image block can be used.In the embodiment of the present invention, can directly using the depth value of fuzzy parameter as image block, can use the depth value of function expression as image block of fuzzy parameter, the embodiment of the present invention does not limit the concrete form of this function expression yet.
Owing to can be determined the fuzzy parameter of each image block by step 103, and then the depth value of each image block can be obtained, therefrom can get the minimum depth value of image.For the arbitrary image block in image, using the fuzzy parameter of this image block as its depth value, relatively the minimum depth value of this depth value and image compares, whether the difference of both judgements is not less than setting threshold value, if difference is not less than setting threshold value, then show that this image block is the darker region of the degree of depth in image (or, be called background/background region), determine that it is not again focal zone (foreground area); If above-mentioned difference is less than setting threshold value, then shows that this image block is the foreground area of image, it can be used as focal zone again, to make it more outstanding in the picture after image procossing.
In actual conditions, can the depth value of image block whether be also again focal zone relative to the amount of the amplitude that deepens of minimum depth value to weigh this image block.Such as, in image, the minimum value of the depth of field is d
1, the maximal value of the depth of field is d
2if the depth value of image block is at interval [d
1, d
1+ t* (d
2-d
1)] in scope, then this image block is within focal zone again, otherwise outside focal zone again, wherein, wherein t is setting ratio value, 0<t<1.
Optionally, in the embodiment of the present invention, after step 103, with reference to Fig. 4, also comprise the steps: step 106: using the fuzzy core parameter of fuzzy parameter as image block, deblurring process corresponding to fuzzy core parameter is carried out to image block.
Concrete, fuzzy core parameter can be the σ of Gaussian Blur core, in the embodiment of the present invention, using the fuzzy parameter of image block as fuzzy core parameter corresponding to image block, Gaussian Blur nuclear operator can be obtained according to fuzzy core parameter, and then utilize this Gaussian Blur nuclear operator to carry out deblurring process to image block, namely, utilize Gaussian Blur nuclear operator and image block to carry out de-convolution operation, after process, the sharpness of image block will improve, and realize the enhancing of picture quality.
Owing to utilizing the fuzzy core parameter of image block to carry out deblurring operation to image, make the not only picture quality enhancing of the image block after deblurring operational processes, and image change is natural, smooth.
Optionally, in the embodiment of the present invention, after step 103, with reference to Fig. 5, can also comprise the steps:
Step 107: according to fuzzy parameter, synthesis corresponds to the fuzzy graph of image, and the pixel value corresponding to the region of image block in fuzzy graph is fuzzy parameter;
Step 108: according to fuzzy graph, determines the depth value of image.
Concrete, in step 107, after the fuzzy parameter obtaining each image block in image block, fuzzy parameter is collectively referred to as the fuzzy graph of original image, synthesis mode is: the pixel value replacing this image block in original image with the fuzzy parameter of image block.
In step 108, because fuzzy parameter can the depth value of token image block, in the embodiment of the present invention, the pixel value (that is, the fuzzy parameter of image block) of each pixel in fuzzy graph is synthesized the depth value of the image characterizing original image depth of field feature.Concrete synthesis mode can be: the weighted mean asking each pixel in blurred picture, such as, ask root mean square; Also can be averaging computing for calmodulin binding domain CaM feature, that is, the weighting factor for zones of different in different fuzzy graph is different, and such as, the weighting factor of the pixel of fuzzy graph middle section is larger.
By the fuzzy graph of composograph, the depth value of image is obtained again according to fuzzy graph, because the fuzzy parameter of this depth value based on all image blocks of original image generates, can reflect entire image main body (or, the prospect things of image) the depth of field, contribute to the spatial positional information knowing objects in images, supplementary can be provided to image procossing such as the splicing of image, registration, three-dimensional modelings.
Based on identical technical conceive, the embodiment of the present invention additionally provides a kind of device 200 determining image blurring degree, and with reference to Fig. 6, the structural schematic block diagram of device 200, comprising:
Sparse coding module 201, for carrying out sparse coding according to sparse dictionary to the image block of image, thus obtains the sparse vector for representing image block;
First determination module 202, for the sparse vector of image block obtained according to sparse coding module 201, determines the sparse features of image block;
Second determination module 203, for the sparse features obtained according to the first determination module 202, determine the fuzzy parameter of image block, fuzzy parameter represents the fog-level of image block.
Optionally, in the embodiment of the present invention, the second determination module 203 for: according to the corresponding relation between sparse features and fuzzy parameter, and the first determination module 202 obtain sparse features, determine fuzzy parameter.
Optionally, in the embodiment of the present invention, the second determination module 203 for: by solve following equation calculate fuzzy parameter:
Wherein, f is sparse features, and σ is fuzzy parameter, and a, b, c, d, e are preset value.
Optionally, in the embodiment of the present invention, the first determination module 202 for zero norm of the sparse vector of computed image block, using zero norm as sparse features.
Optionally, in the embodiment of the present invention, device 200 also comprises:
Fuzzy Processing module 204, for judging image block whether in focal zone again; And when image block is not in focal zone again, Fuzzy Processing corresponding to fuzzy parameter is carried out to image block.
Optionally, in the embodiment of the present invention, Fuzzy Processing module 204 for: obtain the minimum depth value of image; Using the depth value of fuzzy parameter as image block, judge whether the difference that the depth value of image block deducts minimum depth value is not less than setting threshold value; If the difference that the depth value of image block deducts minimum depth value is not less than setting threshold value, then determines image block not in focal zone again, Fuzzy Processing corresponding to fuzzy parameter is carried out to image block.
Optionally, in the embodiment of the present invention, device 200 also comprises:
Deblurring module 205, for using the fuzzy core parameter of fuzzy parameter as image block, carries out deblurring process corresponding to fuzzy core parameter to image block.
Optionally, in the embodiment of the present invention, device 200 also comprises:
3rd determination module 206, for according to fuzzy parameter, synthesize the fuzzy graph corresponding to image, the pixel value corresponding to the region of image block in fuzzy graph is fuzzy parameter; And according to fuzzy graph, determine the depth value of image.
Determine that the method for image blurring degree is based on the aspect of two under same inventive concept in device 200 in the present embodiment and the embodiment of the present invention, detailed description is done to the implementation process of method above, so those skilled in the art can according to the structure of the aforementioned device 200 understood with being described clearly in the present embodiment and implementation process, succinct in order to instructions, has just repeated no more at this.
Based on identical inventive concept, the embodiment of the present invention additionally provides a kind of electronic equipment 300, and with reference to Fig. 7, electronic equipment 300 comprises: processor 301 and storer 302.
Wherein, storer 302 is for storing instruction.
Processor 301, for the instruction in execute store 302, with in the process performing instruction, realizes operating as follows: carry out sparse coding according to sparse dictionary to the image block of image, thus obtain the sparse vector for representing image block; According to the sparse vector of image block, determine the sparse features of image block; According to sparse features, determine the fuzzy parameter of image block, fuzzy parameter represents the fog-level of image block.
Optionally, electronic equipment 300 also comprises bus 303, for realizing the electrical connection of processor 301 and storer 302.
Optionally, processor 301 for: according to sparse features, determine the fuzzy parameter of image block, comprising:
According to the corresponding relation between sparse features and fuzzy parameter, and the sparse features obtained, determine fuzzy parameter.
Optionally, processor 301 for: according to sparse features, determine the fuzzy parameter of image block, be specially: by solve following equation calculate fuzzy parameter:
Wherein, f is sparse features, and σ is fuzzy parameter, and a, b, c, d, e are preset value.
Optionally, processor 301 for: according to the sparse vector of image block, determine the sparse features of image block, be specially: zero norm of the sparse vector of computed image block, using zero norm as sparse features.
Optionally, processor 301 after the fuzzy parameter for determining image block, also for: judge image block whether in focal zone again; When image block is not in focal zone again, Fuzzy Processing corresponding to fuzzy parameter is carried out to image block.
Optionally, processor 301 for: judge image block whether in focal zone again, be specially: the minimum depth value obtaining image; Using the depth value of fuzzy parameter as image block, judge whether the difference that the depth value of image block deducts minimum depth value is not less than setting threshold value; If the difference that the depth value of image block deducts minimum depth value is not less than setting threshold value, then determine image block not in focal zone again.
Optionally, processor 301 for: after determining the fuzzy parameter of image block, also for: using the fuzzy core parameter of fuzzy parameter as image block, deblurring process corresponding to fuzzy core parameter is carried out to image block.
Optionally, processor 301 for: after determining the fuzzy parameter of image block, also for: according to fuzzy parameter, synthesis corresponds to the fuzzy graph of image, and the pixel value corresponding to the region of image block in fuzzy graph is fuzzy parameter; According to fuzzy graph, determine the depth value of image.
Optionally, electronic equipment 300 also comprises: image acquisition units 304, for obtaining image.
Optionally, storer 302 is random access memory, after electronic equipment 300 enters normal operating condition, runs application and operating system in random access memory.
Optionally, electronic equipment 300 also comprises: ROM (read-only memory), when needs run electronic equipment 300, is started by the Basic Input or Output System (BIOS) guidance system be solidificated in ROM (read-only memory), guides electronic equipment 300 to enter normal operating condition.
Determine that the method for image blurring degree is based on the aspect of two under same inventive concept in electronic equipment 300 in the present embodiment and the embodiment of the present invention, detailed description is done to the implementation process of method above, so those skilled in the art can according to the structure of the aforementioned electronic equipment 300 understood with being described clearly in the present embodiment and implementation process, succinct in order to instructions, has just repeated no more at this.
The one or more technical schemes provided in the embodiment of the present invention, at least have following technique effect or advantage:
By utilizing sparse dictionary to encode to image block, obtain the sparse vector of image block, and then obtain the sparse features of image, and according to the fuzzy parameter of sparse features determination image block, and then know the fog-level of image block.The significant difference of the fog-level of the clear area of image block and image is not relied on due to said method, can in the inapparent situation of fog-level difference of the clear area of image block and image, determine the fuzzy parameter of image block, reliable valuation is carried out to the fog-level of image block, implementation is simple, result accurately and reliably those skilled in the art should be understood, embodiments of the invention can be provided as method, system or computer program.Therefore, the present invention can adopt the form of complete hardware embodiment, completely software implementation or the embodiment in conjunction with software and hardware aspect.And the present invention can adopt in one or more form wherein including the upper computer program implemented of computer-usable storage medium (including but not limited to magnetic disk memory, CD-ROM, optical memory etc.) of computer usable program code.
The present invention describes with reference to according to the process flow diagram of the method for the embodiment of the present invention, equipment (system) and computer program and/or block scheme.Should understand can by the combination of the flow process in each flow process in computer program instructions realization flow figure and/or block scheme and/or square frame and process flow diagram and/or block scheme and/or square frame.These computer program instructions can being provided to the processor of multi-purpose computer, special purpose computer, Embedded Processor or other programmable data processing device to produce a machine, making the instruction performed by the processor of computing machine or other programmable data processing device produce device for realizing the function of specifying in process flow diagram flow process or multiple flow process and/or block scheme square frame or multiple square frame.
These computer program instructions also can be stored in can in the computer-readable memory that works in a specific way of vectoring computer or other programmable data processing device, the instruction making to be stored in this computer-readable memory produces the manufacture comprising command device, and this command device realizes the function of specifying in process flow diagram flow process or multiple flow process and/or block scheme square frame or multiple square frame.
Although describe the preferred embodiments of the present invention, those skilled in the art once obtain the basic creative concept of cicada, then can make other change and amendment to these embodiments.So claims are intended to be interpreted as comprising preferred embodiment and falling into all changes and the amendment of the scope of the invention.
Obviously, those skilled in the art can carry out various change and modification to the present invention and not depart from the spirit and scope of the present invention.Like this, if these amendments of the present invention and modification belong within the scope of the claims in the present invention and equivalent technologies thereof, then the present invention is also intended to comprise these change and modification.
Claims (16)
1. determine a method for image blurring degree, it is characterized in that, comprising:
According to sparse dictionary, sparse coding is carried out to the image block of described image, thus obtain the sparse vector for representing described image block;
According to the sparse vector of described image block, determine the sparse features of described image block;
According to described sparse features, determine the fuzzy parameter of described image block, described fuzzy parameter represents the fog-level of described image block.
2. the method for claim 1, is characterized in that, described according to described sparse features, determines the fuzzy parameter of described image block, comprising:
According to the corresponding relation between described sparse features and described fuzzy parameter, and described sparse features, determine described fuzzy parameter.
3. method as claimed in claim 1 or 2, is characterized in that, described according to described sparse features, determines the fuzzy parameter of described image block, comprising:
Described fuzzy parameter is determined by solving following equation:
Wherein, f is described sparse features, and σ is described fuzzy parameter, and a, b, c, d, e are preset value.
4. the method as described in any one of claims 1 to 3, is characterized in that, the described sparse vector according to described image block, determines the sparse features of described image block, comprising:
Calculate zero norm of the sparse vector of described image block, using described zero norm as described sparse features.
5. the method as described in any one of Claims 1-4, is characterized in that, described determine the fuzzy parameter of described image block after, described method also comprises:
Judge described image block whether in focal zone again;
When described image block is not in described focal zone again, Fuzzy Processing corresponding to described fuzzy parameter is carried out to described image block.
6. method as claimed in claim 5, is characterized in that, describedly judges described image block whether in focal zone again, comprising:
Obtain the minimum depth value of described image;
Using the depth value of described fuzzy parameter as described image block, judge whether the difference that the depth value of described image block deducts described minimum depth value is not less than setting threshold value;
If the difference that the depth value of described image block deducts described minimum depth value is not less than described setting threshold value, then determine described image block not in described focal zone again.
7. the method as described in any one of Claims 1-4, is characterized in that, described determine the fuzzy parameter of described image block after, described method also comprises:
Using the fuzzy core parameter of described fuzzy parameter as described image block, deblurring process corresponding to described fuzzy core parameter is carried out to described image block.
8. the method as described in any one of Claims 1-4, is characterized in that, described determine the fuzzy parameter of described image block after, described method also comprises:
According to described fuzzy parameter, synthesis corresponds to the fuzzy graph of described image, and the pixel value corresponding to the region of described image block in described fuzzy graph is described fuzzy parameter;
According to described fuzzy graph, determine the depth value of described image.
9. determine a device for image blurring degree, it is characterized in that, comprising:
Sparse coding module, for carrying out sparse coding according to sparse dictionary to the image block of described image, thus obtains the sparse vector for representing described image block;
First determination module, for the sparse vector of the described image block according to described sparse coding module acquisition, determines the sparse features of described image block;
Second determination module, for the described sparse features obtained according to described first determination module, determine the fuzzy parameter of described image block, described fuzzy parameter represents the fog-level of described image block.
10. device as claimed in claim 9, it is characterized in that, described second determination module is used for: according to the corresponding relation between described sparse features and described fuzzy parameter, and the described sparse features that described first determination module obtains, and determines described fuzzy parameter.
11. devices as described in claim 9 or 10, it is characterized in that, described second determination module is used for: determine described fuzzy parameter by solving following equation:
Wherein, f is described sparse features, and σ is described fuzzy parameter, and a, b, c, d, e are preset value.
12. devices as described in any one of claim 9 to 11, it is characterized in that, described first determination module is used for: zero norm calculating the sparse vector of described image block, using described zero norm as described sparse features.
13. devices as described in any one of claim 9 to 12, it is characterized in that, described device also comprises:
Fuzzy Processing module, for judging described image block whether in focal zone again; And when described image block is not in described focal zone again, Fuzzy Processing corresponding to described fuzzy parameter that described second determination module determines is carried out to described image block.
14. devices as claimed in claim 13, it is characterized in that, described Fuzzy Processing module is used for: the minimum depth value obtaining described image; Using the depth value of described fuzzy parameter as described image block, judge whether the difference that the depth value of described image block deducts described minimum depth value is not less than setting threshold value; If the difference that the depth value of described image block deducts described minimum depth value is not less than described setting threshold value, then determines described image block not in described focal zone again, Fuzzy Processing corresponding to described fuzzy parameter is carried out to described image block.
15. devices as described in any one of claim 9 to 12, it is characterized in that, described device also comprises:
Deblurring module, for using the fuzzy core parameter of described fuzzy parameter as described image block, carries out deblurring process corresponding to described fuzzy core parameter to described image block.
16. devices according to any one of claim 9-11, is characterized in that, also comprise:
3rd determination module, for the described fuzzy parameter determined according to described second determination module, synthesis corresponds to the fuzzy graph of described image, and the pixel value corresponding to the region of described image block in described fuzzy graph is described fuzzy parameter; And according to described fuzzy graph, determine the depth value of described image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510376667.6A CN105096316A (en) | 2015-06-30 | 2015-06-30 | Method and device for determining image blurriness |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510376667.6A CN105096316A (en) | 2015-06-30 | 2015-06-30 | Method and device for determining image blurriness |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105096316A true CN105096316A (en) | 2015-11-25 |
Family
ID=54576656
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510376667.6A Pending CN105096316A (en) | 2015-06-30 | 2015-06-30 | Method and device for determining image blurriness |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105096316A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107403414A (en) * | 2017-07-14 | 2017-11-28 | 华中科技大学 | A kind of image area selecting method and system for being beneficial to fuzzy kernel estimates |
| CN115880300A (en) * | 2023-03-03 | 2023-03-31 | 北京网智易通科技有限公司 | Image blur detection method, image blur detection device, electronic equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2007035144A1 (en) * | 2005-09-22 | 2007-03-29 | Obigo Ab | Method and arrangement for digital filtering of images |
| CN102842110A (en) * | 2011-06-20 | 2012-12-26 | 富士胶片株式会社 | Image processing device and image processing method |
| CN103049906A (en) * | 2012-12-07 | 2013-04-17 | 清华大学深圳研究生院 | Image depth extraction method |
| CN104463775A (en) * | 2014-10-31 | 2015-03-25 | 小米科技有限责任公司 | Device and method for achieving depth-of-field effect of image |
-
2015
- 2015-06-30 CN CN201510376667.6A patent/CN105096316A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2007035144A1 (en) * | 2005-09-22 | 2007-03-29 | Obigo Ab | Method and arrangement for digital filtering of images |
| CN102842110A (en) * | 2011-06-20 | 2012-12-26 | 富士胶片株式会社 | Image processing device and image processing method |
| CN103049906A (en) * | 2012-12-07 | 2013-04-17 | 清华大学深圳研究生院 | Image depth extraction method |
| CN104463775A (en) * | 2014-10-31 | 2015-03-25 | 小米科技有限责任公司 | Device and method for achieving depth-of-field effect of image |
Non-Patent Citations (1)
| Title |
|---|
| 张涛 等: "基于稀疏表示的图像模糊度评价方法", 《计算机工程》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107403414A (en) * | 2017-07-14 | 2017-11-28 | 华中科技大学 | A kind of image area selecting method and system for being beneficial to fuzzy kernel estimates |
| CN107403414B (en) * | 2017-07-14 | 2018-11-02 | 华中科技大学 | A kind of image area selecting method and system being conducive to fuzzy kernel estimates |
| WO2019010932A1 (en) * | 2017-07-14 | 2019-01-17 | 华中科技大学 | Image region selection method and system favorable for fuzzy kernel estimation |
| US11093778B2 (en) | 2017-07-14 | 2021-08-17 | Huazhong University Of Science And Technology | Method and system for selecting image region that facilitates blur kernel estimation |
| CN115880300A (en) * | 2023-03-03 | 2023-03-31 | 北京网智易通科技有限公司 | Image blur detection method, image blur detection device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Bansal et al. | Blur image detection using Laplacian operator and Open-CV | |
| Dobrosotskaya et al. | A wavelet-Laplace variational technique for image deconvolution and inpainting | |
| CN109961444B (en) | Image processing method and device and electronic equipment | |
| US8705881B2 (en) | Motion deblurring for text images | |
| US20180122051A1 (en) | Method and device for image haze removal | |
| CN105740876B (en) | A kind of image pre-processing method and device | |
| CN105335947A (en) | Image de-noising method and image de-noising apparatus | |
| CN106920222B (en) | A kind of image smoothing method and device | |
| US20170109873A1 (en) | Image enhancement using self-examples and external examples | |
| CN102693527A (en) | Method and apparatus for performing a blur rendering process on an image | |
| CN110675334A (en) | Image enhancement method and device | |
| Gao et al. | Detail preserved single image dehazing algorithm based on airlight refinement | |
| CN111598796A (en) | Image processing method and device, electronic device and storage medium | |
| Thanki et al. | Digital image processing using SCILAB | |
| US9396522B2 (en) | Scalable massive parallelization of overlapping patch aggregation | |
| CN101739670B (en) | Non-local mean space domain time varying image filtering method | |
| CN104346782A (en) | Method and device for defogging single image | |
| CN105096316A (en) | Method and device for determining image blurriness | |
| Ma et al. | Mixed noise removal algorithm combining adaptive directional weighted mean filter and improved adaptive anisotropic diffusion model | |
| CN104200438A (en) | Multi-level infrared image detail enhancement processing method and processing device thereof | |
| Sun et al. | Edge guidance filtering for structure extraction | |
| Li et al. | Deep scale-aware image smoothing | |
| Liu et al. | Adaptive sparse norm and nonlocal total variation methods for image smoothing | |
| CN115393868B (en) | Text detection method, device, electronic equipment and storage medium | |
| CN111311610A (en) | Image segmentation method and terminal 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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20151125 |