CN112200247A - Image processing system and method based on multi-dimensional image mapping - Google Patents
Image processing system and method based on multi-dimensional image mapping Download PDFInfo
- Publication number
- CN112200247A CN112200247A CN202011084916.1A CN202011084916A CN112200247A CN 112200247 A CN112200247 A CN 112200247A CN 202011084916 A CN202011084916 A CN 202011084916A CN 112200247 A CN112200247 A CN 112200247A
- Authority
- CN
- China
- Prior art keywords
- dimension
- image
- mapping
- value
- center
- 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
- 238000013507 mapping Methods 0.000 title claims abstract description 159
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000012545 processing Methods 0.000 title claims abstract description 43
- 238000009826 distribution Methods 0.000 claims abstract description 29
- 238000007781 pre-processing Methods 0.000 claims abstract description 4
- 238000007906 compression Methods 0.000 claims description 36
- 230000006835 compression Effects 0.000 claims description 33
- 238000010606 normalization Methods 0.000 claims description 16
- 238000004364 calculation method Methods 0.000 claims description 13
- 238000004458 analytical method Methods 0.000 claims description 12
- 230000014509 gene expression Effects 0.000 claims description 12
- 238000010191 image analysis Methods 0.000 claims description 9
- 239000011159 matrix material Substances 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 238000000354 decomposition reaction Methods 0.000 abstract description 4
- 230000002829 reductive effect Effects 0.000 abstract description 4
- 239000000523 sample Substances 0.000 description 17
- 230000002441 reversible effect Effects 0.000 description 9
- 238000000926 separation method Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 230000002427 irreversible effect Effects 0.000 description 3
- 238000005070 sampling Methods 0.000 description 3
- 238000013144 data compression Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000003672 processing method Methods 0.000 description 2
- 230000003595 spectral effect Effects 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 230000002123 temporal effect Effects 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 108010001267 Protein Subunits Proteins 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 230000000670 limiting effect Effects 0.000 description 1
- 230000001681 protective effect Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/50—Extraction of image or video features by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
Abstract
The invention belongs to the technical field of image processing, and particularly relates to an image processing system and method based on multi-dimensional image mapping. The system comprises: and the image preprocessing unit is configured to perform image gray level histogram processing on the image to be processed based on a plurality of preset different gray value ranges, and respectively obtain corresponding gray level histogram characteristic distributions in the 27 different gray value ranges. The method comprises the steps of firstly, conducting histogram processing on an image to be processed based on different gray values to obtain gray level histogram characteristic distribution, conducting multi-dimensional mapping on the basis to achieve multi-dimensional decomposition of the image, further processing the image in each dimension, and finally conducting merging processing on the image, so that the redundancy of similar images is greatly reduced, and storage space and system resources are saved.
Description
Technical Field
The invention belongs to the technical field of image processing, and particularly relates to an image processing system and method based on multi-dimensional image mapping.
Background
Image data can be compressed because of the redundancy present in the data. The redundancy of image data is mainly represented by: spatial redundancy due to correlation between adjacent pixels in the image; temporal redundancy caused by correlation between different frames in the image sequence; spectral redundancy due to the correlation of different color planes or spectral bands. The goal of data compression is to reduce the number of bits required to represent the data by removing these data redundancies. Since the amount of image data is enormous, it is very difficult to store, transmit, and process the image data, and thus compression of the image data is very important.
The information age brings "information explosion" and greatly increases the data volume, so that the data needs to be effectively compressed regardless of transmission or storage. In the remote sensing technology, various space probes adopt a compression coding technology to send acquired huge information back to the ground.
Image compression is the application of data compression techniques to digital images with the goal of reducing redundant information in the image data so that the data can be stored and transmitted in a more efficient format.
Patent No. CN2010102134318A discloses an image compression apparatus for performing compression processing on an input image having a plurality of color components, comprising: a first image separation unit that separates the input image into a reversible compression image in which reversible compression is performed and an irreversible compression image in which irreversible compression is performed, based on pixel attribute information indicating to which of a plurality of regions including a character region and a photograph region each a pixel constituting the input image belongs; a color determination unit that determines color information to be used when the reversible compression image is further separated; a second image separation unit that separates the reversible compression image separated by the first image separation unit into a first reversible compression image including the color information determined by the color determination unit and one or more second reversible compression images other than the first reversible compression image; and an image compression unit that performs different compression processes on the irreversible compression image separated by the first image separation unit, the first reversible compression image separated by the second image separation unit, and the second reversible compression image, respectively. Although the image is compressed under the condition of ensuring the image quality, the compression rate is still not high, and the image redundancy is not effectively eliminated.
Disclosure of Invention
In view of the above, the present invention provides an image processing system and method based on multi-dimensional image mapping, which first perform histogram processing on an image to be processed based on different gray values to obtain gray level histogram characteristic distribution, and perform multi-dimensional mapping on the basis to realize multi-dimensional decomposition of the image, so as to process the image in each dimension, and finally perform merging processing on the image, thereby greatly reducing redundancy of similar images and saving storage space and system resources.
In order to achieve the purpose, the technical scheme of the invention is realized as follows:
an image processing system based on multi-dimensional image mapping, the system comprising:
the image preprocessing unit is configured to perform image gray level histogram processing on an image to be processed based on a plurality of preset different gray value ranges, and obtain corresponding gray level histogram characteristic distributions under the 27 different gray value ranges respectively;
the image mapping unit is configured to perform multi-dimensional mapping on an image to be processed based on the obtained gray level histogram characteristic distribution to obtain a mapping value in each dimension, and establish a multi-dimensional mapping set for each dimension based on the obtained mapping value;
the image processing unit comprises a plurality of image processors, the number of the image processors is the same as the number of the dimensions used for carrying out multi-dimensional mapping, each image processor calculates the normalized approximation degree between the mapping value of one dimension of the dimension under the multi-dimensional mapping and the mapping value of the adjacent dimension under the corresponding dimension aiming at each dimension, after all the image processors finish the calculation of the normalized approximation degree under the corresponding dimension, the number of the dimension mapping values with the statistical approximation degree higher than a set threshold value is counted, and the number is used as the image affinity degree of the dimension and the adjacent dimension; the adjacent dimension is another dimension adjacent to the storage position of the dimension;
the image compression unit is used for classifying two dimensions of which the image affinity exceeds a set threshold value as a class; after classification is completed, calculating to obtain a normalization center according to the dimensions of all the classes, and merging the images of the same class based on the calculated normalization center so as to reduce redundancy among the images of the same class.
Further, the image mapping unit includes: the characteristic extraction subunit is configured to decompose the image to be processed based on the obtained gray level histogram characteristic distribution to obtain input image characteristic expressions of the image to be processed in 27 scales, 9 depths and 3 types, wherein each scale corresponds to 3 depths and 1 type; and the mapping subunit is configured to perform image mapping dimension mapping on the image feature expression under each scale, each depth and each type to obtain mapping values of the image to be processed under 27 dimensions, and establish a multi-dimension mapping set based on the obtained mapping values.
Further, the step of calculating the normalization center by the image compression unit includes: the total number of categories is marked as G, and the normalization center is obtained by calculation through the following formula
(ii) a Wherein G is the total number of categories, c is the normalized number, N is the total number of samples, UGRepresenting a membership matrix, V, in the G-th dimensionGDenotes the normalized center in the G-th dimension, XGRepresenting the G-th dimensionally small normalized sample,
representing the center point of the ith class in the G-th dimension, d is the dimension number of the sample, xj，GDenotes the j sample point, μ, in the G dimensionij，GRepresenting the membership degree of the jth sample belonging to the ith class under the G dimensionality, wherein m is an adjustment coefficient and must satisfy m<1: according to the established normalization center, the normalization center is established,
is the normalized center.
Further, the image analysis unit performs line normalization analysis based on the established normalization center, and includes the following steps: calculating the distance between each dimension and the normalization center; obtaining a coordinate point of the dimensionality according to the calculated distance; forming a set by all the obtained coordinate points, and using the set as a dimensional coordinate point set; and carrying out normalization analysis on the obtained dimensional coordinate point combination.
Further, the image processor, calculating a normalized approximation between the mapping value of one of the dimensions under the multi-dimensional mapping and the mapping value of the adjacent dimension under the corresponding dimension, comprises the following steps: setting the distance transformation function of each dimension mapping value as:
wherein d (p, q) represents a set of euclidean distances of each dimension mapping value, and p and q represent the abscissa and ordinate of each dimension, respectively; dimension mapping value Da(p) ordinate, D, representing dimension mapping valueb(q) abscissa representing dimension mapping value, IbRepresenting a range value of the abscissa, wherein the value range is { 2-10 }; h isbAnd the range value of the vertical coordinate is { 3-12 }.
Further, when classifying, the image analysis unit traverses the whole dimension, and a plurality of dimensions exist in the same category.
A method of image processing based on multi-dimensional image mapping, the method performing the steps of:
step 1: performing image gray level histogram processing on an image to be processed based on a plurality of preset different gray value ranges, and respectively obtaining corresponding gray level histogram characteristic distributions under 27 different gray value ranges;
step 2: carrying out multi-dimensional mapping on the image to be processed based on the obtained gray level histogram characteristic distribution to obtain a mapping value under each dimension, and establishing a multi-dimensional mapping set aiming at each dimension based on the obtained mapping value;
and step 3: establishing a plurality of image processors, wherein the number of the image processors is the same as the number of the dimensions used for carrying out multi-dimensional mapping, each image processor calculates the normalized approximation degree between the mapping value of one dimension of the dimension under the multi-dimensional mapping and the mapping value of the adjacent dimension under the corresponding dimension aiming at each dimension, and after all the image processors finish the calculation of the normalized approximation degree under the corresponding dimension, the number of the dimension mapping values with the approximation degree higher than a set threshold value is counted and is used as the image affinity degree of the dimension and the adjacent dimension; the adjacent dimension is another dimension adjacent to the storage position of the dimension;
and 4, step 4: classifying two dimensions of which the image affinity exceeds a set threshold value to serve as a category; after classification is completed, calculating to obtain a normalization center according to the dimensions of all the classes, and merging the images of the same class based on the calculated normalization center so as to reduce redundancy among the images of the same class.
Further, step 2 comprises: decomposing an image to be processed based on the obtained gray level histogram characteristic distribution to obtain the characteristic expressions of the image to be processed in 27 scales, 9 depths and 3 types of input images, wherein each scale corresponds to 3 depths and 1 type; and the mapping subunit is configured to perform image mapping dimension mapping on the image feature expression under each scale, each depth and each type to obtain mapping values of the image to be processed under 27 dimensions, and establish a multi-dimension mapping set based on the obtained mapping values.
Further, the step of calculating the normalization center includes the following steps: the total number of categories is marked as G, and the normalization center is obtained by calculation through the following formula
(ii) a Wherein G is the total number of categories, c is the normalized number, N is the total number of samples, UGRepresenting a membership matrix, V, in the G-th dimensionGDenotes the normalized center in the G-th dimension, XGRepresenting the G-th dimensionally small normalized sample,
representing the center point of the ith class in the G-th dimension, d is the dimension number of the sample, xj，GDenotes the j sample point, μ, in the G dimensionij，GRepresenting the membership degree of the jth sample belonging to the ith class under the G dimensionality, wherein m is an adjustment coefficient and must satisfy m<1: according to the established normalization center, the normalization center is established,
is the normalized center.
Further, the image analysis unit performs line normalization analysis based on the established normalization center, and includes the following steps: calculating the distance between each dimension and the normalization center; obtaining a coordinate point of the dimensionality according to the calculated distance; forming a set by all the obtained coordinate points, and using the set as a dimensional coordinate point set; and carrying out normalization analysis on the obtained dimensional coordinate point combination.
The image processing system and method based on multi-dimensional image mapping have the following beneficial effects: the method comprises the steps of firstly, conducting histogram processing on an image to be processed based on different gray values to obtain gray level histogram characteristic distribution, conducting multi-dimensional mapping on the basis to achieve multi-dimensional decomposition of the image, further processing the image in each dimension, and finally conducting merging processing on the image, so that the redundancy of similar images is greatly reduced, and storage space and system resources are saved. The method is mainly realized by the following steps: 1. gray scale histogram property: the method carries out image gray level histogram processing on the image to be processed based on a plurality of preset different gray level value ranges, respectively obtains corresponding gray level histogram characteristic distribution under 27 different gray level value ranges, reflects the image characteristics under the gray level value ranges on the gray level histogram characteristics under each gray level value range, and carries out multi-dimensional mapping based on the gray level histogram characteristics, thereby reflecting the image characteristics in more detail and accurately; 2. multidimensional mapping: the method comprises the steps of carrying out multi-dimensional mapping on an image to be processed based on obtained gray level histogram characteristic distribution to obtain a mapping value under each dimension, and establishing a multi-dimensional mapping set aiming at each dimension based on the obtained mapping value; the process can obtain the mapping values of the images under multiple dimensions, and based on the mapping values, image similar parts in a single image or among multiple images can be obtained, and on the basis, the similar parts can be processed in the compression process to reduce the redundancy of the compressed images, so that the storage efficiency and the compression efficiency are improved; 3. the algorithms of the normalization center and the normalization approximation degree are different from the prior art, and can provide support for subsequent compression processing.
Drawings
FIG. 1 is a schematic system diagram of an image processing system based on multi-dimensional image mapping according to an embodiment of the present invention;
fig. 2 is a flowchart illustrating a method of an image processing method based on multi-dimensional image mapping according to an embodiment of the present invention.
Detailed Description
The method of the present invention will be described in further detail below with reference to the accompanying drawings and embodiments of the invention.
Example 1
As shown in fig. 1, an image processing system based on multi-dimensional image mapping, the system comprising:
the image preprocessing unit is configured to perform image gray level histogram processing on an image to be processed based on a plurality of preset different gray value ranges, and obtain corresponding gray level histogram characteristic distributions under the 27 different gray value ranges respectively;
the image mapping unit is configured to perform multi-dimensional mapping on an image to be processed based on the obtained gray level histogram characteristic distribution to obtain a mapping value in each dimension, and establish a multi-dimensional mapping set for each dimension based on the obtained mapping value;
the image processing unit comprises a plurality of image processors, the number of the image processors is the same as the number of the dimensions used for carrying out multi-dimensional mapping, each image processor calculates the normalized approximation degree between the mapping value of one dimension of the dimension under the multi-dimensional mapping and the mapping value of the adjacent dimension under the corresponding dimension aiming at each dimension, after all the image processors finish the calculation of the normalized approximation degree under the corresponding dimension, the number of the dimension mapping values with the statistical approximation degree higher than a set threshold value is counted, and the number is used as the image affinity degree of the dimension and the adjacent dimension; the adjacent dimension is another dimension adjacent to the storage position of the dimension;
the image compression unit is used for classifying two dimensions of which the image affinity exceeds a set threshold value as a class; after classification is completed, calculating to obtain a normalization center according to the dimensions of all the classes, and merging the images of the same class based on the calculated normalization center so as to reduce redundancy among the images of the same class.
By adopting the technical scheme, the image to be processed is subjected to histogram processing on the basis of different gray values to obtain gray level histogram characteristic distribution, on the basis, multi-dimensional mapping is performed to realize multi-dimensional decomposition of the image, the image can be processed in each dimension, and finally the image is subjected to merging processing, so that the redundancy of similar images is greatly reduced, and the storage space and system resources are saved. The method is mainly realized by the following steps: 1. gray scale histogram property: the method carries out image gray level histogram processing on the image to be processed based on a plurality of preset different gray level value ranges, respectively obtains corresponding gray level histogram characteristic distribution under 27 different gray level value ranges, reflects the image characteristics under the gray level value ranges on the gray level histogram characteristics under each gray level value range, and carries out multi-dimensional mapping based on the gray level histogram characteristics, thereby reflecting the image characteristics in more detail and accurately; 2. multidimensional mapping: the method comprises the steps of carrying out multi-dimensional mapping on an image to be processed based on obtained gray level histogram characteristic distribution to obtain a mapping value under each dimension, and establishing a multi-dimensional mapping set aiming at each dimension based on the obtained mapping value; the process can obtain the mapping values of the images under multiple dimensions, and based on the mapping values, image similar parts in a single image or among multiple images can be obtained, and on the basis, the similar parts can be processed in the compression process to reduce the redundancy of the compressed images, so that the storage efficiency and the compression efficiency are improved; 3. the algorithms of the normalization center and the normalization approximation degree are different from the prior art, and can provide support for subsequent compression processing.
Example 2
On the basis of the above embodiment, the image mapping unit includes: the characteristic extraction subunit is configured to decompose the image to be processed based on the obtained gray level histogram characteristic distribution to obtain input image characteristic expressions of the image to be processed in 27 scales, 9 depths and 3 types, wherein each scale corresponds to 3 depths and 1 type; and the mapping subunit is configured to perform image mapping dimension mapping on the image feature expression under each scale, each depth and each type to obtain mapping values of the image to be processed under 27 dimensions, and establish a multi-dimension mapping set based on the obtained mapping values.
Example 3
On the basis of the above embodiment, the step of calculating the normalization center by the image compression unit includes the following steps: the total number of categories is marked as G, and the normalization center is obtained by calculation through the following formula
(ii) a Wherein G is the total number of categories, c is the normalized number, N is the total number of samples, UGDenotes the G thMembership matrix, V, under individual dimensionGDenotes the normalized center in the G-th dimension, XGRepresenting the G-th dimensionally small normalized sample,
representing the center point of the ith class in the G-th dimension, d is the dimension number of the sample, xj，GDenotes the j sample point, μ, in the G dimensionij，GRepresenting the membership degree of the jth sample belonging to the ith class under the G dimensionality, wherein m is an adjustment coefficient and must satisfy m<1: according to the established normalization center, the normalization center is established,
is the normalized center.
Specifically, normalization is a dimensionless processing means, which makes the absolute value of the physical system value become a certain relative value relationship. Simplifying the calculation and reducing the magnitude. [1] For example, after each frequency value in the filter is normalized by the cutoff frequency, the frequency is a relative value of the cutoff frequency, and there is no dimension. After the impedance is normalized by the internal resistance of the power supply, each impedance becomes a relative impedance value, and the dimension of ohm does not exist. After all kinds of operation are finished, all the inverse normalization is recovered. A nyquist frequency, defined as one-half of the sampling frequency, is often used in signal processing toolsets, and the cutoff frequency in both order selection and design of the filter is normalized using the nyquist frequency. For example, for a system with a sampling frequency of 500hz, the normalized frequency of 400hz is 400/500 ═ 0.8, and the normalized frequency range is between [0,1 ]. If the normalized frequency is converted to angular frequency, the normalized frequency is multiplied by 2 x pi, if the normalized frequency is converted to hz, the normalized frequency is multiplied by half the sampling frequency.
Example 4
On the basis of the above embodiment, the image analysis unit performs line normalization analysis based on the established normalization center, and includes the following steps: calculating the distance between each dimension and the normalization center; obtaining a coordinate point of the dimensionality according to the calculated distance; forming a set by all the obtained coordinate points, and using the set as a dimensional coordinate point set; and carrying out normalization analysis on the obtained dimensional coordinate point combination.
Example 5
On the basis of the above embodiment, the image processor, calculating the normalized approximation between the mapping value of one dimension of the dimension under the multi-dimension mapping and the mapping value of the adjacent dimension under the corresponding dimension, comprises the following steps: setting the distance transformation function of each dimension mapping value as:
Specifically, the redundancy of the digital image mainly takes the following forms: spatial redundancy, temporal redundancy, visual redundancy, information entropy redundancy, structural redundancy, and knowledge redundancy. Spatial redundancy: there is much redundancy due to strong correlation between adjacent pixels within an image. Time redundancy: redundancy due to correlation between different frames in a sequence of video images. Visual redundancy: refers to the portion of the image information that the human eye cannot perceive or is insensitive. Information entropy redundancy: also called coding redundancy, a redundancy exists in a picture if the number of bits used per pixel on average in the picture is greater than the information entropy of the picture, and this redundancy is called information entropy redundancy. Structural redundancy: meaning that there is a strong texture or self-similarity in the image. Knowledge redundancy: meaning that some images also contain information about some verification knowledge.
Example 6
On the basis of the above embodiment, when performing classification, the image analysis unit traverses the entire dimension, and multiple dimensions exist in the same category.
Example 7
As shown in fig. 2, an image processing method based on multi-dimensional image mapping performs the following steps:
step 1: performing image gray level histogram processing on an image to be processed based on a plurality of preset different gray value ranges, and respectively obtaining corresponding gray level histogram characteristic distributions under 27 different gray value ranges;
step 2: carrying out multi-dimensional mapping on the image to be processed based on the obtained gray level histogram characteristic distribution to obtain a mapping value under each dimension, and establishing a multi-dimensional mapping set aiming at each dimension based on the obtained mapping value;
and step 3: establishing a plurality of image processors, wherein the number of the image processors is the same as the number of the dimensions used for carrying out multi-dimensional mapping, each image processor calculates the normalized approximation degree between the mapping value of one dimension of the dimension under the multi-dimensional mapping and the mapping value of the adjacent dimension under the corresponding dimension aiming at each dimension, and after all the image processors finish the calculation of the normalized approximation degree under the corresponding dimension, the number of the dimension mapping values with the approximation degree higher than a set threshold value is counted and is used as the image affinity degree of the dimension and the adjacent dimension; the adjacent dimension is another dimension adjacent to the storage position of the dimension;
and 4, step 4: classifying two dimensions of which the image affinity exceeds a set threshold value to serve as a category; after classification is completed, calculating to obtain a normalization center according to the dimensions of all the classes, and merging the images of the same class based on the calculated normalization center so as to reduce redundancy among the images of the same class.
Specifically, the basic principle of lossless compression is that the same color information needs to be saved only once. The software that compresses the image first determines which regions of the image are the same and which are different. Images including duplicate data (e.g., blue sky) may be compressed and only the start and end points of the blue sky may need to be recorded. However, the blue color may also have a different shade, and the sky may sometimes be masked by trees, mountains, or other objects, which need to be recorded separately. In essence, the lossless compression method can remove some of the duplicated data, greatly reducing the size of the image to be saved on the disk. However, lossless compression methods do not reduce the memory footprint of the image because the software fills in the missing pixels with the appropriate color information when the image is read from disk. If the memory capacity of the image is to be reduced, a lossy compression method must be used.
Example 8
On the basis of the above embodiment, step 2 includes: decomposing an image to be processed based on the obtained gray level histogram characteristic distribution to obtain the characteristic expressions of the image to be processed in 27 scales, 9 depths and 3 types of input images, wherein each scale corresponds to 3 depths and 1 type; and the mapping subunit is configured to perform image mapping dimension mapping on the image feature expression under each scale, each depth and each type to obtain mapping values of the image to be processed under 27 dimensions, and establish a multi-dimension mapping set based on the obtained mapping values.
Specifically, the image to be processed is subjected to image gray level histogram processing based on a plurality of preset different gray level value ranges, corresponding gray level histogram characteristic distributions in the 27 different gray level value ranges are respectively obtained, the image characteristics in the gray level range are reflected according to the gray level histogram characteristics in each gray level value range, and then multi-dimensional mapping is performed based on the gray level histogram characteristics, so that the image characteristics can be reflected in more detail and accurately.
Example 9
On the basis of the previous embodiment, the step of calculating the normalization center includes the following steps: the total number of categories is marked as G, and the normalization center is obtained by calculation through the following formula
(ii) a Wherein G is the total number of categories, c is the normalized number, N is the total number of samples, UGRepresenting a membership matrix, V, in the G-th dimensionGDenotes the normalized center in the G-th dimension, XGRepresenting the G-th dimensionally small normalized sample,
representing the center point of the ith class in the G-th dimension, d is the dimension number of the sample, xj，GDenotes the j sample point, μ, in the G dimensionij，GRepresenting the membership degree of the jth sample belonging to the ith class under the G dimensionality, wherein m is an adjustment coefficient and must satisfy m<1: according to the established normalization center, the normalization center is established,
is the normalized center.
Example 10
On the basis of the above embodiment, the image analysis unit performs line normalization analysis based on the established normalization center, and includes the following steps: calculating the distance between each dimension and the normalization center; obtaining a coordinate point of the dimensionality according to the calculated distance; forming a set by all the obtained coordinate points, and using the set as a dimensional coordinate point set; and carrying out normalization analysis on the obtained dimensional coordinate point combination.
The method comprises the steps of carrying out multi-dimensional mapping on an image to be processed based on obtained gray level histogram characteristic distribution to obtain a mapping value under each dimension, and establishing a multi-dimensional mapping set aiming at each dimension based on the obtained mapping value; the process can obtain the mapping values of the images under multiple dimensions, and based on the mapping values, similar parts of the images in a single image or among multiple images can be obtained.
It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working process and related description of the system described above may refer to the corresponding process in the foregoing method embodiments, and will not be described herein again.
It should be noted that, the system provided in the foregoing embodiment is only illustrated by dividing the functional units, and in practical applications, the functions may be distributed by different functional units according to needs, that is, the units or steps in the embodiments of the present invention are further decomposed or combined, for example, the units in the foregoing embodiment may be combined into one unit, or may be further decomposed into multiple sub-units, so as to complete all or the functions of the units described above. The names of the units and steps involved in the embodiments of the present invention are only for distinguishing the units or steps, and are not to be construed as unduly limiting the present invention.
It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes and related descriptions of the storage device and the processing device described above may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
Those of skill in the art would appreciate that the various illustrative elements, method steps, described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that programs corresponding to the elements, method steps may be located in Random Access Memory (RAM), memory, Read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. To clearly illustrate this interchangeability of electronic hardware and software, various illustrative components and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as electronic hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
The terms "first," "second," and the like are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order.
The terms "comprises," "comprising," or any other similar term are intended to cover a non-exclusive inclusion, such that a process, method, article, or unit/apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or unit/apparatus.
So far, the technical solutions of the present invention have been described in connection with the preferred embodiments shown in the drawings, but it is easily understood by those skilled in the art that the scope of the present invention is obviously not limited to these specific embodiments. Equivalent modifications or substitutions of the related art marks may be made by those skilled in the art without departing from the principle of the present invention, and the technical solutions after such modifications or substitutions will fall within the protective scope of the present invention.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the scope of the present invention.
Claims (10)
1. An image processing system based on multi-dimensional image mapping, the system comprising:
the image preprocessing unit is configured to perform image gray level histogram processing on an image to be processed based on a plurality of preset different gray value ranges, and obtain corresponding gray level histogram characteristic distributions under the 27 different gray value ranges respectively;
the image mapping unit is configured to perform multi-dimensional mapping on an image to be processed based on the obtained gray level histogram characteristic distribution to obtain a mapping value in each dimension, and establish a multi-dimensional mapping set for each dimension based on the obtained mapping value;
the image processing unit comprises a plurality of image processors, the number of the image processors is the same as the number of the dimensions used for carrying out multi-dimensional mapping, each image processor calculates the normalized approximation degree between the mapping value of one dimension of the dimension under the multi-dimensional mapping and the mapping value of the adjacent dimension under the corresponding dimension aiming at each dimension, after all the image processors finish the calculation of the normalized approximation degree under the corresponding dimension, the number of the dimension mapping values with the statistical approximation degree higher than a set threshold value is counted, and the number is used as the image affinity degree of the dimension and the adjacent dimension; the adjacent dimension is another dimension adjacent to the storage position of the dimension;
the image compression unit is used for classifying two dimensions of which the image affinity exceeds a set threshold value as a class; after classification is completed, calculating to obtain a normalization center according to the dimensions of all the classes, and merging the images of the same class based on the calculated normalization center so as to reduce redundancy among the images of the same class.
2. The system of claim 1, wherein the image mapping unit comprises: the characteristic extraction subunit is configured to decompose the image to be processed based on the obtained gray level histogram characteristic distribution to obtain input image characteristic expressions of the image to be processed in 27 scales, 9 depths and 3 types, wherein each scale corresponds to 3 depths and 1 type; and the mapping subunit is configured to perform image mapping dimension mapping on the image feature expression under each scale, each depth and each type to obtain mapping values of the image to be processed under 27 dimensions, and establish a multi-dimension mapping set based on the obtained mapping values.
3. The system of claim 2, wherein the image compression unit calculating the normalized center comprises the steps of: the total number of categories is marked as G, and the normalization center is obtained by calculation through the following formula
(ii) a Wherein G is the total number of categories, c is the normalized number, N is the total number of samples, UGRepresenting a membership matrix, V, in the G-th dimensionGDenotes the normalized center in the G-th dimension, XGRepresenting the G-th dimensionally small normalized sample,
representing the center point of the ith class in the G-th dimension, d is the dimension number of the sample, xj，GDenotes the j sample point, μ, in the G dimensionij，GAnd (3) representing the membership degree of the jth sample belonging to the ith class under the G dimension, wherein m is an adjustment coefficient and must satisfy m < 1: according to the established normalization center, the normalization center is established,
is the normalized center.
4. The system of claim 3, wherein the image analysis unit performs line normalization analysis based on the established normalization center comprising the steps of: calculating the distance between each dimension and the normalization center; obtaining a coordinate point of the dimensionality according to the calculated distance; forming a set by all the obtained coordinate points, and using the set as a dimensional coordinate point set; and carrying out normalization analysis on the obtained dimensional coordinate point combination.
5. The system of claim 4, wherein the image processor, calculating a normalized approximation between the mapped value of one of the dimensions under the multi-dimensional mapping and the mapped value of the adjacent dimension under the corresponding dimension, comprises the steps of: setting the distance transformation function of each dimension mapping value as:
wherein d (p, q) represents eachA set of Euclidean distances of the dimension mapping values, p and q respectively representing the abscissa and the ordinate of each dimension; dimension mapping value Da(p) ordinate, D, representing dimension mapping valueb(q) abscissa representing dimension mapping value, IbRepresenting a range value of the abscissa, wherein the value range is { 2-10 }; h isbAnd the range value of the vertical coordinate is { 3-12 }.
6. The system of claim 5, wherein the image analysis unit, when performing the categorization, will traverse the entire dimension, with multiple dimensions in the same category.
7. A method of image processing based on multi-dimensional image mapping based on the system of one of claims 1 to 6, characterized in that the method performs the following steps:
step 1: performing image gray level histogram processing on an image to be processed based on a plurality of preset different gray value ranges, and respectively obtaining corresponding gray level histogram characteristic distributions under 27 different gray value ranges;
step 2: carrying out multi-dimensional mapping on the image to be processed based on the obtained gray level histogram characteristic distribution to obtain a mapping value under each dimension, and establishing a multi-dimensional mapping set aiming at each dimension based on the obtained mapping value;
and step 3: establishing a plurality of image processors, wherein the number of the image processors is the same as the number of the dimensions used for carrying out multi-dimensional mapping, each image processor calculates the normalized approximation degree between the mapping value of one dimension of the dimension under the multi-dimensional mapping and the mapping value of the adjacent dimension under the corresponding dimension aiming at each dimension, and after all the image processors finish the calculation of the normalized approximation degree under the corresponding dimension, the number of the dimension mapping values with the approximation degree higher than a set threshold value is counted and is used as the image affinity degree of the dimension and the adjacent dimension; the adjacent dimension is another dimension adjacent to the storage position of the dimension;
and 4, step 4: classifying two dimensions of which the image affinity exceeds a set threshold value to serve as a category; after classification is completed, calculating to obtain a normalization center according to the dimensions of all the classes, and merging the images of the same class based on the calculated normalization center so as to reduce redundancy among the images of the same class.
8. The method of claim 7, wherein step 2 comprises: decomposing an image to be processed based on the obtained gray level histogram characteristic distribution to obtain the characteristic expressions of the image to be processed in 27 scales, 9 depths and 3 types of input images, wherein each scale corresponds to 3 depths and 1 type; and the mapping subunit is configured to perform image mapping dimension mapping on the image feature expression under each scale, each depth and each type to obtain mapping values of the image to be processed under 27 dimensions, and establish a multi-dimension mapping set based on the obtained mapping values.
9. The method of claim 8, wherein said calculating a normalization center comprises the steps of: the total number of categories is marked as G, and the normalization center is obtained by calculation through the following formula
(ii) a Wherein G is the total number of categories, c is the normalized number, N is the total number of samples, UGRepresenting a membership matrix, V, in the G-th dimensionGDenotes the normalized center in the G-th dimension, XGRepresenting the G-th dimensionally small normalized sample,
representing the center point of the ith class in the G-th dimension, d is the dimension number of the sample, xj，GRepresents the G thJ sample point in dimension, μij，GAnd (3) representing the membership degree of the jth sample belonging to the ith class under the G dimension, wherein m is an adjustment coefficient and must satisfy m < 1: according to the established normalization center, the normalization center is established,
is the normalized center.
10. The method of claim 9, wherein the image analysis unit performs line normalization analysis based on the established normalization center, comprising the steps of: calculating the distance between each dimension and the normalization center; obtaining a coordinate point of the dimensionality according to the calculated distance; forming a set by all the obtained coordinate points, and using the set as a dimensional coordinate point set; and carrying out normalization analysis on the obtained dimensional coordinate point combination.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011084916.1A CN112200247B (en) | 2020-10-12 | 2020-10-12 | Image processing system and method based on multi-dimensional image mapping |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011084916.1A CN112200247B (en) | 2020-10-12 | 2020-10-12 | Image processing system and method based on multi-dimensional image mapping |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112200247A true CN112200247A (en) | 2021-01-08 |
| CN112200247B CN112200247B (en) | 2021-07-02 |
Family
ID=74014043
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011084916.1A Active CN112200247B (en) | 2020-10-12 | 2020-10-12 | Image processing system and method based on multi-dimensional image mapping |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112200247B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101706871A (en) * | 2009-11-05 | 2010-05-12 | 上海交通大学 | Isometric mapping based facial image recognition method |
| CN101751565A (en) * | 2008-12-10 | 2010-06-23 | 中国科学院自动化研究所 | Method for character identification through fusing binary image and gray level image |
| CN103400154A (en) * | 2013-08-09 | 2013-11-20 | 电子科技大学 | Human body movement recognition method based on surveillance isometric mapping |
| CN103942572A (en) * | 2014-05-07 | 2014-07-23 | 中国标准化研究院 | Method and device for extracting facial expression features based on bidirectional compressed data space dimension reduction |
| US9922432B1 (en) * | 2016-09-02 | 2018-03-20 | Artomatix Ltd. | Systems and methods for providing convolutional neural network based image synthesis using stable and controllable parametric models, a multiscale synthesis framework and novel network architectures |
| US20190138420A1 (en) * | 2017-11-07 | 2019-05-09 | Vmware, Inc. | Methods and systems that efficiently store and analyze multidimensional metric data |
| CN109840584A (en) * | 2017-11-24 | 2019-06-04 | 腾讯科技（深圳）有限公司 | Convolutional neural networks model, data processing method and device |
| CN110300977A (en) * | 2017-02-17 | 2019-10-01 | 考吉森公司 | Method for image procossing and video compress |
| CN110971901A (en) * | 2018-09-29 | 2020-04-07 | 杭州海康威视数字技术股份有限公司 | Convolutional neural network processing method and device |
-
2020
- 2020-10-12 CN CN202011084916.1A patent/CN112200247B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101751565A (en) * | 2008-12-10 | 2010-06-23 | 中国科学院自动化研究所 | Method for character identification through fusing binary image and gray level image |
| CN101706871A (en) * | 2009-11-05 | 2010-05-12 | 上海交通大学 | Isometric mapping based facial image recognition method |
| CN103400154A (en) * | 2013-08-09 | 2013-11-20 | 电子科技大学 | Human body movement recognition method based on surveillance isometric mapping |
| CN103942572A (en) * | 2014-05-07 | 2014-07-23 | 中国标准化研究院 | Method and device for extracting facial expression features based on bidirectional compressed data space dimension reduction |
| US9922432B1 (en) * | 2016-09-02 | 2018-03-20 | Artomatix Ltd. | Systems and methods for providing convolutional neural network based image synthesis using stable and controllable parametric models, a multiscale synthesis framework and novel network architectures |
| CN110300977A (en) * | 2017-02-17 | 2019-10-01 | 考吉森公司 | Method for image procossing and video compress |
| US20190138420A1 (en) * | 2017-11-07 | 2019-05-09 | Vmware, Inc. | Methods and systems that efficiently store and analyze multidimensional metric data |
| CN109840584A (en) * | 2017-11-24 | 2019-06-04 | 腾讯科技（深圳）有限公司 | Convolutional neural networks model, data processing method and device |
| CN110971901A (en) * | 2018-09-29 | 2020-04-07 | 杭州海康威视数字技术股份有限公司 | Convolutional neural network processing method and device |
Non-Patent Citations (3)
| Title |
|---|
| BART GOOSSENS等: ""A split-augmented Lagrangian Algorithm for Spectral Factorization of a set of 2D Directional Filters and Application to the Design of Compact Shearlet Frames"", 《PROCEEDINGS OF SPIE - THE INTERNATIONAL SOCIETY FOR OPTICAL ENGINEERING》 * |
| 李雨妮等: ""一种使用多维度直方图匹配的图像风格迁移算法"", 《计算机与现代化》 * |
| 束鑫等: ""基于最小外接圆补偿机制的形状图像检索"", 《计算机科学》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112200247B (en) | 2021-07-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110147710B (en) | Method and device for processing human face features and storage medium | |
| Kekre et al. | Improved texture feature based image retrieval using Kekre’s fast codebook generation algorithm | |
| Wang et al. | Reduced-reference image quality assessment using a wavelet-domain natural image statistic model | |
| Xue et al. | Reduced reference image quality assessment based on Weibull statistics | |
| CN112001847A (en) | Method for generating high-quality image by relatively generating antagonistic super-resolution reconstruction model | |
| CN110225350B (en) | Natural image compression method based on generation type countermeasure network | |
| CN106851437A (en) | A kind of method for extracting video frequency abstract | |
| CN109872305B (en) | No-reference stereo image quality evaluation method based on quality map generation network | |
| CN110349101A (en) | A kind of multiple dimensioned united Wavelet image denoising method | |
| CN110647813A (en) | Human face real-time detection and identification method based on unmanned aerial vehicle aerial photography | |
| CN114187463A (en) | Electronic archive generation method and device, terminal equipment and storage medium | |
| CN112200247B (en) | Image processing system and method based on multi-dimensional image mapping | |
| CN111209862B (en) | Clustering method, device and medium for face images | |
| CN106101732B (en) | The vector quantization scheme of Fast Compression bloom spectrum signal | |
| CN112699901A (en) | Plant species identification system based on Internet of things | |
| CN109977769B (en) | Method for identifying micro expression in low-resolution environment | |
| CN108230411B (en) | Method and device for detecting tampered image | |
| CN113542771A (en) | Video high-efficiency compression processing method based on content weight | |
| Abd-Alzhra et al. | Lossy image compression using hybrid deep learning autoencoder based on k-mean clustering | |
| CN111639212A (en) | Image retrieval method in mining intelligent video analysis | |
| CN113128525A (en) | Control device and method for desert grassland population patch identification | |
| CN111243046A (en) | Image quality detection method, device, electronic equipment and storage medium | |
| CN115205952B (en) | Online learning image acquisition method and system based on deep learning | |
| CN116152530B (en) | Image difference determining method and device, storage medium and electronic equipment | |
| CN115795370B (en) | Electronic digital information evidence obtaining method and system based on resampling trace |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20210615 Address after: Room 11703b, 17 / F, block B, Xi'an National Digital publishing base, 996 tianguqi Road, Yuhua Street office, high tech Zone, Xi'an City, Shaanxi Province, 710000 Applicant after: Xi'an zetayun Technology Co.,Ltd. Address before: 311404 Chakou village, Dongqiao Town, Fuyang District, Hangzhou City, Zhejiang Province Applicant before: Hangzhou Wuji communication equipment Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP01 | Change in the name or title of a patent holder | ||
| CP01 | Change in the name or title of a patent holder |
Address after: Room 11703b, 17 / F, block B, Xi'an National Digital publishing base, 996 tianguqi Road, Yuhua Street office, high tech Zone, Xi'an City, Shaanxi Province, 710000 Patentee after: Xi'an Ruisi Shuzhi Technology Co.,Ltd. Address before: Room 11703b, 17 / F, block B, Xi'an National Digital publishing base, 996 tianguqi Road, Yuhua Street office, high tech Zone, Xi'an City, Shaanxi Province, 710000 Patentee before: Xi'an zetayun Technology Co.,Ltd. |
|
| PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: Image processing system and method based on multidimensional image mapping Granted publication date: 20210702 Pledgee: Xi'an innovation financing Company limited by guarantee Pledgor: Xi'an Ruisi Shuzhi Technology Co.,Ltd. Registration number: Y2024980006421 |