WO2017190691A1

WO2017190691A1 - Picture compression method and apparatus

Info

Publication number: WO2017190691A1
Application number: PCT/CN2017/083241
Authority: WO
Inventors: 王康; 黄贞
Original assignee: 贵州白山云科技有限公司
Priority date: 2016-05-05
Filing date: 2017-05-05
Publication date: 2017-11-09
Also published as: CN105979265A

Abstract

Provided are a picture compression method and apparatus. The method comprises: acquiring a first picture; cutting the first picture into a plurality of bitmaps; determining a compression coding algorithm for each of the plurality of bitmaps; using the determined compression coding algorithm to compress a corresponding bitmap; and splicing compressed bitmaps into a second picture. By means of the present invention, the problems of low compression ratio and picture distortion caused by compressing a picture using a single compression algorithm are solved, the compression ratio of a picture is increased, and the distortion of picture compression is reduced.

Description

Picture compression method and device

The present application claims the priority of the Chinese Patent Application, filed on May 5, 2016, which is hereby incorporated by reference. .

Technical field

The present invention relates to the field of picture compression, and in particular to a picture compression method and apparatus.

Background technique

In the "Internet +" era, enterprises transmit information to their users through the website. Diversified terminals need to transmit and receive various types of pictures on the website. Common requirements include: users through personal computers (Personal Computer, PC for short) browsers, mobile browsers Access images, access images via a tablet (PAD) or mobile app (APP), upload or download images via mobile phone, receive or transfer images via instant messaging software, and more. In general, the amount of image data is relatively large. In order to save bandwidth traffic, storage space, and shorten image processing time, it is necessary to compress or compress the image before sending and receiving.

At present, in the field of digital image processing technology, the purpose of encoding and compressing pictures is usually to transform and combine picture data according to certain rules, so as to represent as much image information as possible with as few code symbols as possible. Currently, methods for image compression mainly include lossy compression and lossless compression.

Among them, lossy compression removes the sudden change of color in the image, and deletes the sudden change of color in the image, that is, the human brain will use the closest color to the nearby to fill the lost color characteristics. The compression ratio of lossy compression is higher than that of lossless compression and can be used to satisfy storage and delivery requirements; however, lossy compression techniques can affect picture quality, and there will be more or less distortion when restoring original data.

Lossless compression uses design redundancy to remove duplicate data for compression to reduce the size of images saved on disk. Lossless compression does not cause any distortion to fully recover the original data; however, the compression ratio is limited by the statistical redundancy of the data. The lossy compression and lossless compression of the image apply to different types of images. Lossy compression is suitable for pictures with uniform color distribution. The compressed picture generally has higher distortion and higher compression ratio; lossless compression is suitable for pictures with relatively simple color, and the compressed picture generally has high security. Trueness and lower compression ratio.

In the current digital image processing technology, when the client software product compresses the picture, it usually adopts a compression algorithm for compression according to the traditional compression algorithm for any picture content form. Obviously, a single algorithm cannot accommodate all images; a single quality cannot handle the pixel portion of each bitmap of the image. Even if the image compression is implemented according to the existing compression technology, only a picture with low compression ratio and distortion is obtained, thereby affecting the user experience.

An effective solution has not been given in the related art for the problem of low compression rate and picture distortion caused by compressing pictures using a single compression algorithm.

Summary of the invention

The present invention provides a picture compression method and apparatus to solve at least the above problems.

According to an aspect of the present invention, a picture compression method is provided, including: acquiring a first picture; cutting the first picture into a plurality of bitmaps; and determining compression coding of each bitmap in the plurality of bitmaps The algorithm compresses the corresponding bitmap by using a determined compression coding algorithm; and splices the compressed bitmap into a second picture.

Optionally, determining a compression coding algorithm for each bitmap in the plurality of bitmaps includes: determining pixel quality of each bitmap in the plurality of bitmaps; and selecting each bitmap according to the pixel quality Compression coding algorithm.

Optionally, according to the pixel quality, selecting a compression coding algorithm for each bitmap includes: determining whether a pixel quality of each bitmap in the plurality of bitmaps is less than a predetermined threshold; and if the determination result is yes Selecting the compression encoding algorithm of the bitmap in a first set of compression encoding algorithms; otherwise, selecting the compression encoding algorithm of the bitmap in a second set of compression encoding algorithms.

Optionally, the first compression coding algorithm set is a lossy compression coding algorithm set; and the second compression coding algorithm set is a lossless compression coding algorithm set.

Optionally, the set of lossy compression coding algorithms includes at least one of the following: a model coding algorithm, a fractal coding algorithm, and a discrete cosine transform coding algorithm; the lossless compression coding algorithm set includes at least one of the following: integer wavelet fast The hierarchical tree set splitting lossless compression algorithm SSPIHT, the improved adaptive run length coding algorithm, and the improved binary bit level coding algorithm.

Optionally, cutting the first picture into a plurality of bitmaps includes: following the edge detection method The first picture is cut into the plurality of bitmaps.

Optionally, after the compressed multiple bitmaps are spliced into the second picture, the method further includes: storing the second picture; and/or outputting the second picture.

According to another aspect of the present invention, a picture compression apparatus is provided, including: an obtaining module, configured to acquire a first picture; a cutting module, configured to cut the first picture into a plurality of bitmaps; and a determining module, a compression coding algorithm for determining each bitmap in the plurality of bitmaps; a compression module, configured to compress a corresponding bitmap by using a determined compression coding algorithm; and a splicing module, configured to: compress the plurality of bits The picture is stitched into a second picture.

Optionally, the determining module includes: a determining unit, configured to determine a pixel quality of each bitmap in the multiple bitmaps; and a selecting unit, configured to select compression coding of each bitmap according to the pixel quality algorithm.

Optionally, the selecting unit includes: a determining subunit, configured to determine whether the pixel quality of each bitmap in the multiple bitmaps is less than a predetermined threshold; and selecting a subunit for determining that the result is In the case of the first compression coding algorithm set, the compression coding algorithm of the bitmap is selected; otherwise, the compression coding algorithm of the bitmap is selected in the second compression coding algorithm set.

According to the present invention, a method of acquiring a first picture, cutting a first picture into a plurality of bitmaps, determining a compression coding algorithm for each bitmap in the plurality of bitmaps, and compressing the corresponding bitmap by using the determined compression coding algorithm is adopted. The invention solves the problem of low compression rate and picture distortion caused by compressing pictures by using a single compression algorithm, improves the compression ratio of the picture, and reduces the distortion of the picture compression.

DRAWINGS

The drawings described herein are intended to provide a further understanding of the invention, and are intended to be a part of the invention. In the drawing:

1 is a flowchart of a picture compression method according to an embodiment of the present invention;

2 is a schematic structural diagram of a picture compression apparatus according to an embodiment of the present invention;

3 is a flow chart of a method for cutting a compression stitching algorithm based on a picture bitmap according to a preferred embodiment of the present invention;

4 is a diagram of an apparatus for cutting a compression stitching algorithm based on a picture bitmap according to a preferred embodiment of the present invention. Schematic.

detailed description

The invention will be described in detail below with reference to the drawings in conjunction with the embodiments. It should be noted that the embodiments in the present application and the features in the embodiments may be combined with each other without conflict.

It is to be understood that the terms "first", "second" and the like in the specification and claims of the present invention are used to distinguish similar objects, and are not necessarily used to describe a particular order or order.

A picture compression method is provided in this embodiment. FIG. 1 is a flowchart of a picture compression method according to an embodiment of the present invention. As shown in FIG. 1, the process includes the following steps:

Step S101, acquiring a first picture;

Step S102, cutting the first picture into a plurality of bitmaps;

Step S103, determining a compression coding algorithm for each bitmap in the plurality of bitmaps;

Step S104, compressing the corresponding bitmap by using a determined compression coding algorithm;

Step S105, the compressed and encoded plurality of bitmaps are spliced into a second picture.

After the first picture is cut into multiple bitmaps, the multiple bitmaps obtained by the cutting are respectively compression-encoded, and different compression coding algorithms are used for different cutting parts of the first picture; The picture size of the picture does not change. Therefore, multiple bitmaps can be re-spliced at the end, so that the second picture compressed by different compression coding algorithms can be obtained. It can be seen that through the above steps, the problem of low compression rate and picture distortion caused by compressing the picture by using a single compression algorithm is solved, the compression ratio of the picture is improved, and the distortion of the picture compression is reduced.

In step S103, a compression coding algorithm may be determined according to the pixel quality of each bitmap, for example, determining the pixel quality of each bitmap in the plurality of bitmaps; and selecting a compression coding algorithm for each bitmap according to the pixel quality. Among them, the pixel quality includes but is not limited to: comprehensive quantitative indicators such as definition, contrast, color shift, signal to noise ratio and number of dead points. The compression algorithm is selected according to different pixel qualities, so that the cut portions of different pixel qualities can adopt different compression coding algorithms, thereby obtaining different compression ratios and distortion levels.

Optionally, according to the pixel quality, the compression coding algorithm for selecting each bitmap may adopt the following manner: determining whether the pixel quality of each bitmap in the multiple bitmaps is less than a predetermined threshold; if the determination result is yes, Selecting the compression coding algorithm of the bitmap in the first set of compression coding algorithms; otherwise, in the A compression coding algorithm that selects this bitmap in a set of two compression coding algorithms. Comparing the quantized pixel quality with a predetermined threshold, and selecting a corresponding compression coding algorithm in the first compression coding algorithm set or the second compression coding algorithm set according to different comparison results, the predetermined threshold may be 5%.

Optionally, the foregoing first compression coding algorithm set is a set of lossy compression coding algorithms, including but not limited to: a model coding algorithm, a fractal coding algorithm, and a discrete cosine transform coding algorithm.

Optionally, the foregoing second compression coding algorithm set is a lossless compression coding algorithm set, including but not limited to: integer wavelet fast multi-level tree set split lossless compression algorithm SSPIHT, improved adaptive run length coding algorithm, improved binary bit level coding algorithm .

Optionally, when the first picture is cut into multiple bitmaps in step S102, the cutting may be performed according to a predetermined rule. For example, the simplest cutting method is to perform geometric cutting, that is, the image is directly cut into equal or unequal Multiple geometric figures. In the embodiment of the present invention, the first picture is preferably cut into a plurality of bitmaps according to an edge detection method. The method for detecting the edge may be any one of the methods disclosed in the prior art, and is not limited in the embodiment of the present invention. The edge detection method is used to cut the first picture, and the similarly-shaped part of the first picture can be cut into a bitmap, and the part of the first picture with a large difference in nature is cut, which is advantageous for adopting different compression coding. Compression coding of the algorithm.

Optionally, after the compressed multiple bitmaps are spliced into the second picture, the second picture may be stored; and/or the second picture is output.

Through the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, by hardware, but in many cases, the former is A better implementation. Based on such understanding, the technical solution of the present invention, which is essential or contributes to the prior art, may be embodied in the form of a software product stored in a storage medium (such as ROM/RAM, disk, The optical disc includes a number of instructions for causing a terminal device (which may be a cell phone, a computer, a server, or a network device, etc.) to perform the methods of various embodiments of the present invention.

In the embodiment, a picture compression device is also provided, which is used to implement the above-mentioned embodiments and preferred embodiments, and has not been described again. As used below, the term "module" may implement a combination of software and/or hardware of a predetermined function. Although the apparatus described in the following embodiments is preferably implemented in software, hardware, or a combination of software and hardware, is also possible and contemplated.

2 is a schematic structural diagram of a picture compression apparatus according to an embodiment of the present invention. As shown in FIG. 2, the picture compression apparatus includes: an acquisition module 20, a cutting module 22, a determination module 24, a compression module 26, and a splicing module 28, where The obtaining module 20 is configured to acquire a first picture; the cutting module 22 is coupled to the obtaining module 20 for cutting the first picture into a plurality of bitmaps; the determining module 24 is coupled to the cutting module 22 for determining a plurality of positions a compression coding algorithm for each bitmap in the figure; a compression module 26 coupled to the determination module 24 for compressing the corresponding bitmap using the determined compression coding algorithm; a splicing module 28 coupled to the compression module 26 for compressing Multiple bitmaps are stitched into a second picture.

Optionally, the determining module 24 includes: a determining unit, configured to determine a pixel quality of each bitmap in the plurality of bitmaps; and a selecting unit coupled to the determining unit, configured to select a compression encoding of each bitmap according to the pixel quality algorithm.

Optionally, the selecting unit includes: a determining subunit, configured to determine whether a pixel quality of each bitmap in the plurality of bitmaps is less than a predetermined threshold; and selecting a subunit, coupled to the determining subunit, for determining that the result is In the case of the first compression coding algorithm set, the compression coding algorithm of the bitmap is selected; otherwise, the compression coding algorithm of the bitmap is selected in the second compression coding algorithm set.

Optionally, the first compression coding algorithm set is a lossy compression coding algorithm set; the second compression coding algorithm set is a lossless compression coding algorithm set.

Optionally, the set of lossy compression coding algorithms includes, but is not limited to, at least one of the following: a model coding algorithm, a fractal coding algorithm, and a discrete cosine transform coding algorithm; the lossless compression coding algorithm set includes but is not limited to at least one of the following: integer wavelet Fast multi-level tree set splitting lossless compression algorithm SSPIHT, improved adaptive run length coding algorithm, improved binary bit level coding algorithm.

Optionally, the cutting module 22 is configured to cut the first picture into a plurality of bitmaps according to an edge detection method.

Optionally, the device may further include: a storage module coupled to the splicing module 28 for storing the second picture; and/or an output module coupled to the storage splicing module 28 or the storage module for outputting the second picture.

It should be noted that each of the above modules may be implemented by software or hardware. For the latter, the foregoing may be implemented by, but not limited to, the foregoing modules are all located in the same processor; or, the modules are located in multiple In the processor.

Embodiments of the present invention also provide a software for performing the technical solutions described in the above embodiments and preferred embodiments.

Embodiments of the present invention also provide a storage medium. In this embodiment, the foregoing storage medium may be Program code that is set to store for performing the following steps:

Step S101, acquiring a first picture;

Step S102, cutting the first picture into a plurality of bitmaps;

Optionally, in the embodiment, the foregoing storage medium may include, but is not limited to, a USB flash drive, a Read-Only Memory (ROM), and a Random Access Memory (RAM). A variety of media that can store program code, such as a hard disk, a disk, or an optical disk.

For example, the specific examples in this embodiment may refer to the examples described in the foregoing embodiments and the optional embodiments, and details are not described herein again.

In order to make the description of the embodiments of the present invention more clear, the following description and description are given in conjunction with the preferred embodiments.

The object of the preferred embodiment of the present invention is to provide a method and a device for cutting and compressing a stitching algorithm based on a picture bitmap, and according to the size of the picture bitmap, after determining the pixel quality of the bitmap, intelligently selecting an appropriate compression algorithm, and then performing the compression. The picture is stitched together to effectively implement high fidelity compression of the picture.

A preferred embodiment of the present invention provides a method for cutting a compression stitching algorithm based on a picture bitmap. 3 is a flow chart of a method for cutting a compression stitching algorithm based on a picture bitmap according to a preferred embodiment of the present invention. As shown in FIG. 3, the method includes the following steps:

Step S1001: input the original picture, determine the original picture size, and cut the original picture into multiple bitmaps of different sizes according to the edge detection method;

Optionally, the segmentation picture is implemented by detecting edge edges of different regions in the image. The edge is the most basic feature of the picture, which is the result of discontinuity or mutation of the local features of the picture. The zero point information of the extreme or second derivative of the first derivative of the picture is used to provide the basic basis of the edge point for processing the gray value, color and texture mutation; for constructing the difference sensitive to the gray step change of the image The operator performs picture bitmap segmentation. The preferred embodiment of the present invention adopts a local picture function based method and a multi-scale method and a boundary curve fitting method, and utilizes a first order differential operator: a Robert algorithm, a Sobel algorithm, and a second order differential calculation. Sub: Laplace algorithm and Kirsh implement the segmentation of picture bitmaps. For example: resolution is 1920*1400 The picture is cut into 60 small pictures. After performing edge detection on the first picture, a binary image is obtained, and the small picture is divided according to the edge point density in the binary image. For example, the area where the edge point density is less than the preset density threshold is divided into a small picture, and the edge point density is greater than the pre- The area where the density threshold is set is divided into a small picture. The size of each small picture can be set to be the same or different. In the method, a unit area may be set, and the unit area may be a rectangle of a preset size, typically a square, and the first picture is evenly divided into a plurality of small pictures by a unit area, and the edge point density of each unit area is calculated. An area including a plurality of connected edge points having a density less than a preset density threshold merges the plurality of areas into a rectangular area as a small picture, if a rectangular area includes a plurality of connected edge points having a density greater than The area of the preset density threshold combines the multiple areas into one rectangular area as a small picture.

Step S1002, determining the pixel quality of each of the cut bitmaps, that is, detecting the comprehensive quality of various aspects such as sharpness, contrast, color shift, signal to noise ratio, and number of dead points;

Optionally, a determination is made as to the quality of each cut bitmap pixel. Determining the pixel quality of each bitmap in the plurality of bitmaps includes: performing at least one of the following operations on each bitmap and calculating a pixel quality of each bitmap according to the operation result: a pixel brightness value of the statistical bitmap, The edge contour value of the statistical bitmap, the gray average of the statistical bitmap, the gray variance of the statistical bitmap, and the gray histogram of the statistical bitmap. In a preferred embodiment of the present invention, the pixel luminance values of the divided picture bitmaps are counted, and the edge contour values of the bitmaps transmitted in step S1002 are statistically recorded. And statistically divide the grayscale average and grayscale variance of each picture bitmap. Calculate the function of the bitmap gray histogram, that is, the number of pixels with the gray level in the picture, reflecting the frequency of occurrence of different gray levels in the picture bitmap; used to check whether the picture is reasonable to use all the allowed gray The range of degrees; and the selection of the boundary threshold and the calculation of the integrated optical density; the preferred embodiment of the present invention utilizes a grayscale histogram function in conjunction with the pixel luminance value and the edge contour value to determine the picture bitmap local seek quality threshold preferably 5%.

Step S1003, if the pixel quality of the cut small bitmap is lower than the threshold of 5%, determine which lossy compression encoding algorithm is suitable for each pixel of the cut small bitmap;

To analyze the pixel quality of each of the cut small bitmaps below the threshold of 5%, it is determined which lossy compression coding algorithm is suitable. Lossy compression is to reduce the redundancy of picture pixel correlation, and use the coding method to remove these redundancy, which is used to reduce redundant compressed data. That is, for the divided bitmap, between the adjacent pixels of the same row of the picture, the corresponding pixels of the adjacent frames of the moving picture have a relatively strong correlation between the adjacent pixels, and the correlation is taken out or reduced. Is to remove or reduce the picture bitmap information The redundancy in the middle thus achieves lossy compression of the segmented bitmap pixels.

Step S1004, selecting an optimal lossy compression coding algorithm, performing lossy compression on each of the cut small bitmap pixels;

The lossy compression algorithm includes, but is not limited to, one of model coding, fractal coding, discrete cosine transform coding, or any combination thereof. Each compression algorithm has its own characteristics and is suitable for images in different formats. For example, for a quality threshold higher than 5%, the color richness is high, and it is suitable to be processed by a lossy algorithm. The lossy compression of the picture will lose some information, but because of the high quality of the bitmap, the color richness is high, all the lost information will not affect the user's vision, and the compression ratio is low. In the preferred embodiment, the lossy compression algorithm is based on Discrete Cosine Transform (DCT). In other embodiments, any other algorithm that can implement lossy compression may be used. An enumeration.

Step S1005: If the pixel quality of the small bitmap after the cutting is less than the threshold value of 5%, it is determined whether the pixel of each of the cut small bitmaps is suitable for which lossless compression encoding algorithm is used;

To analyze whether the pixel quality of each of the cut small bitmaps is below the threshold of 5%, then it is determined which lossless compression coding algorithm is suitable. Lossless coding utilizes the statistical properties of the source to remove the intrinsic correlation and change the non-uniformity of the probability distribution, thereby realizing the compression of image information. The lossless compression accuracy is high. Due to the statistical probability based method and the dictionary-based technology, the image is lost with less information when recompressing, and thus has higher precision. The compression ratio of lossless compression is small, that is, the connection between the pixels of the picture is almost completely preserved, and the image is more accurate, so the compression ratio is small and the space is large; the lossless compression is reversible without losing the original information content.

Step S1006, selecting an optimal lossless compression coding algorithm, performing lossless compression on each of the cut small bitmap pixels;

The lossless compression algorithm includes, but is not limited to, one of the integer wavelet fast SPIHT lossless compression algorithm SSPIHT (Speed SPIHT), the improved adaptive run length coding, the improved binary (bit level) lossless image compression method, or any combination thereof. For example, rich in color and uneven distribution, suitable for integer image wavelet fast SPIHT lossless compression algorithm SSPIHT (Speed SPIHT) for bitmap image operation with quality threshold less than 5%, this method combined with integer wavelet transform decomposition, when the threshold is less than or equal to 4 For the problem of the increase of unimportant coefficient, the bit plane coding method is introduced to optimize the SPIHT algorithm to improve the compression ratio and compression speed.

Step S1007, finding a global optimal bitmap suture according to the algorithm, and compressing the bitmap image after the splicing and dividing;

Optionally, based on the bitmap splicing of the image, the global optimal suture is found after the cutting is completed by using step S1001, and then the final synthesis is realized by Poisson fusion technology. The weight calculation method based on the gradient direction histogram statistics is performed by calculating the gradient direction histogram of each pixel region of the bitmap after compression, and then combining the gradient intensity and direction of the pixel points to perform weight calculation, thereby improving Based on the instability of the weight calculation method of color intensity or gradient strength, a better graph cutting suture search is completed. The Poisson fusion method combined with overlapping transitions is divided into two unknown regions by overlapping regions, and the smoothing is defined respectively. The transitioned boundary values are used to implement the processing of image differences.

In step S1007, in order to enable the second picture to be successfully decoded and displayed, stitching the compressed bitmap into the second picture includes: writing size information of each bitmap in the header file of the second picture, The location information in the first picture and the compression coding algorithm information used. According to the header file in the second picture, the first picture can be obtained and displayed normally.

Step S1008, storing a compressed new picture;

In step S1009, the compressed new picture is output.

Compared with the prior art, the method provided by the preferred embodiment of the present invention can determine the quality bitmap quality threshold for the small bitmap pixels after the picture segmentation, thereby intelligently determining whether it belongs to lossy compression or lossless compression, and then selecting corresponding The compression algorithm, the number of different pixels reflects the richness of the color of the picture, and then splicing the compressed picture small bitmap to complete the uniform splicing of the picture color distribution, thereby ensuring effective compression of each bitmap pixel of the cut picture. Stitching ensures the high quality effect of the picture after cutting and compressing.

A preferred embodiment of the present invention also provides an apparatus for cutting a compression stitching algorithm based on a picture bitmap. 4 is a schematic structural diagram of an apparatus for cutting a compression splicing algorithm based on a picture bitmap according to a preferred embodiment of the present invention. As shown in FIG. 4, the apparatus includes:

a picture cutting device 4001, configured to cut an original picture and cut into a plurality of bitmaps of different sizes according to an edge detection method;

The picture quality analysis device 4002 is configured to determine the bitmap after cutting, determine the pixel quality of each cut bitmap and compare the threshold value by 5%, and select an appropriate compression algorithm for different bitmaps according to the comparison result;

The picture compression device 4003 includes a lossy compression unit 40031 for lossy compression of the post-cut bitmap, and a lossless compression unit 40032 for lossless compression of the post-cut bitmap, wherein:

If the pixel quality of the small bitmap after cutting is less than 5% of the threshold, determine which lossy compression coding algorithm is suitable for each pixel of the small bitmap after cutting, and then select the optimal lossy compression coding algorithm for each After cutting, the small bitmap pixels are subjected to lossy compression;

If the pixel quality of the small bitmap after cutting is lower than the threshold value of 5%, it is determined whether the pixel of each small bitmap after cutting is suitable for which lossless compression encoding algorithm, and then the optimal lossless compression encoding algorithm is selected, for each After cutting, the small bitmap pixels are subjected to lossless compression;

The picture splicing device 4004 is configured to perform global optimal bitmap stitching and splicing on the cut and compressed plurality of bitmaps to complete splicing a new picture.

The workflow of the above device will be described below.

First, the picture cutting device 4001 obtains a divided picture realized by edge accumulation of different areas in the image. Among them, the edge is the most basic feature of the picture, which is the result of the discontinuity or mutation of the local features of the picture. Therefore, next, the picture cutting device 4001 uses the extreme value of the first derivative of the picture or the zero point information of the second derivative to provide the basic basis of the edge point for processing the mutation of the gray value, the color and the texture, and is used for constructing The difference operator sensitive to the gray level step change of the image is used to perform image bitmap segmentation. At the same time, the local image function based method and the multi-scale method and the boundary curve fitting method are adopted in the present invention, and the first-order differential operator is utilized: Robert algorithm, Sobel algorithm and second-order differential operator: Laplace algorithm and Kirsh realize the segmentation of picture bitmap.

Next, the picture quality analysis device 4002 determines the quality of each of the cut bitmap pixels, wherein:

The picture quality analysis device 4002 counts the pixel brightness values of the divided picture bitmaps, and statistically records the edge contour values of the bitmaps transmitted from the picture cutting device 4001. And the picture quality analysis device 4002 counts the gray average value and the gray scale variance of each picture bitmap after the division. The picture quality analysis device 4002 calculates a function of the bitmap gray histogram, that is, the number of pixels having the gray level in the picture, reflecting the frequency of occurrence of different gray levels in the picture bitmap; for checking whether the picture is rationally utilized All allowed gray level ranges, and selecting boundary thresholds and calculating integrated optical density; the picture quality analysis device 4002 uses the gray histogram function to combine the pixel brightness values and the edge contour values to comprehensively determine the picture bitmap local search quality threshold 5%.

If it is analyzed that the pixel quality of each small bitmap after cutting is below 5% of the threshold, a lossy compression coding algorithm is suitable. In this case, the picture compression device 4003 processes the picture using lossy compression. Among them, lossy compression is to reduce the redundancy of picture pixel correlation, and use the coding method to delete these redundancy, which is used to reduce redundant compressed data. That is, for the divided bitmap, between the adjacent pixels of the same row of the picture, the corresponding pixels of the adjacent frames of the moving picture have a relatively strong correlation between the adjacent pixels, and the correlation is taken out or reduced. It is to remove or reduce the redundancy in the picture bitmap information to achieve lossy compression of the segmented bitmap pixels. Lossless compression algorithms include, but are not limited to, one of model based coding, fractal coding, discrete cosine transform coding, or any combination thereof. Each compression algorithm has its own characteristics and is suitable for images in different formats. In the preferred embodiment, the lossy compression algorithm is based on Discrete Cosine Transform (DCT). In other embodiments, any other algorithm that can implement lossy compression may be used. .

If it is analyzed that the pixel quality of each of the post-cut small bitmaps is below 5% of the threshold, the picture compression device 4003 employs a lossless compression coding algorithm. Lossless coding utilizes the statistical properties of the source to remove the intrinsic correlation and change the non-uniformity of the probability distribution, thereby realizing the compression of image information. The lossless compression accuracy is high. Due to the statistical probability based method and the dictionary-based technology, the image is lost with less information when recompressing, and thus has higher precision. The compression ratio of the lossless compression of the picture compression device 4003 is small, that is, the connection between the picture pixels is almost completely preserved, so that the image is more accurate, so the compression ratio is smaller and the space is larger; the lossless compression of the device is reversible. Without losing the original information content. In this embodiment, the integer wavelet fast SPIHT lossless compression algorithm SSPIHT (Speed SPIHT). In other embodiments, any other algorithm that can implement lossless compression may also be used, and will not be enumerated here.

The picture splicing device 4004 is configured to perform global optimal bitmap stitching and splicing on the cut and compressed plurality of bitmaps to complete splicing a new picture. The Poisson fusion method using overlapping transitions is used to implement image fusion processing after graph cutting. By dividing each overlapping area into two adjacent unknown areas, the pixel intensity of the adjacent sides takes the average unknown pixel intensity on the two source pictures, and the pixel intensity of the side opposite the adjacent side is Take only the pixel intensity on one image to achieve an overlapping transition. The unknown region is divided and then merged, the Poisson fusion parameters are initialized, and the unknown region is Poisson fusion solution, and finally the fusion transition is completed.

The above-mentioned preferred embodiments effectively solve the problems in the prior art. The device device provided by the preferred embodiment of the present invention can compress and compress the spliced content according to the picture, and select corresponding points. Cut, compress, and splicing algorithms to efficiently perform compression and maintain high fidelity of the image. In addition, the original image is cut into a plurality of bitmaps of different sizes according to the edge detection method, and then the quality of each cut bitmap pixel is judged, and the gray level histogram function is used to combine the pixel brightness value and the edge contour value to comprehensively determine the picture position. The map seeks the quality threshold, and finally adopts the Poisson fusion method of overlapping transition to realize the image fusion processing after the graph cut. Therefore, based on the three methods, the accuracy of the selection and splicing of the image cutting compression algorithm is ensured, thereby ensuring the processed image has a good access effect.

In summary, according to the above embodiments and preferred embodiments of the present invention, bitmap segmentation is performed according to the size of the picture, and after determining the pixel quality of each bitmap, it is accurately determined that each of the cut small bitmap pixels adopts appropriate compression. The algorithm can ensure different images adopt different suitable algorithms to achieve high fidelity effect of the image; in addition, the post-cutting compression algorithm selects the optimal lossy compression or lossless compression according to the pixel quality judgment of the divided image, and selects different compressions. The algorithm is based on the pixel quality of the small picture bitmap after cutting. The number of different pixels reflects the richness of the picture color, and then the compressed picture small bitmap is spliced to complete the uniform splicing of the picture color distribution. Based on these three aspects, the image content is accurately optimized, and the precise compression and optimization of the compression algorithm and the optimization of the splicing algorithm ensure that the image after cutting and splicing achieves a high quality user experience.

It will be apparent to those skilled in the art that the various modules or steps of the present invention described above can be implemented by a general-purpose computing device that can be centralized on a single computing device or distributed across a network of multiple computing devices. Alternatively, they may be implemented by program code executable by the computing device such that they may be stored in the storage device by the computing device and, in some cases, may be different from the order herein. The steps shown or described are performed, or they are separately fabricated into individual integrated circuit modules, or a plurality of modules or steps thereof are fabricated as a single integrated circuit module. Thus, the invention is not limited to any specific combination of hardware and software.

The above description is only the preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes can be made to the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and scope of the present invention are intended to be included within the scope of the present invention.

Industrial applicability

The invention can accurately optimize the picture content, and selects the compression algorithm and optimizes the splicing algorithm by precise cutting, thereby ensuring the quality of the user experience after cutting and splicing the picture.

Claims

A picture compression method, comprising:

Get the first picture;

Cutting the first picture into a plurality of bitmaps;

Determining a compression coding algorithm for each bitmap in the plurality of bitmaps;

Compressing a corresponding bitmap by using a determined compression coding algorithm;

The compressed bitmap is stitched into a second picture.
The method according to claim 1, wherein the compression encoding algorithm for determining each bitmap in the plurality of bitmaps comprises:

Determining a pixel quality of each bitmap in the plurality of bitmaps;

A compression encoding algorithm for each bitmap is selected based on the pixel quality.
The method according to claim 2, wherein the compression encoding algorithm for selecting each bitmap according to the pixel quality comprises:

Determining whether a pixel quality of each bitmap in the plurality of bitmaps is less than a predetermined threshold;

If the determination result is yes, the compression coding algorithm of the bitmap is selected in the first compression coding algorithm set; otherwise, the compression coding algorithm of the bitmap is selected in the second compression coding algorithm set .
The method of claim 3 wherein:

The first compression coding algorithm set is a lossy compression coding algorithm set;

The second compression coding algorithm set is a lossless compression coding algorithm set.
The method of claim 4 wherein:

The set of lossy compression coding algorithms includes at least one of the following: a model coding algorithm, a fractal coding algorithm, and a discrete cosine transform coding algorithm;

The lossless compression coding algorithm set includes at least one of the following: an integer wavelet fast multi-level tree set split lossless compression algorithm SSPIHT, an improved adaptive run length coding algorithm, and an improved binary bit level coding algorithm.
The method of claim 1 wherein said first picture is cut into a plurality of sheets Bitmaps include:

The first picture is cut into the plurality of bitmaps according to an edge detection method.
The method according to claim 1, wherein the determining the pixel quality of each of the plurality of bitmaps comprises: performing at least one of the following operations on each of the bitmaps and calculating each of the operations according to the operation result Pixel quality of the bitmap: the pixel brightness value of the statistical bitmap, the edge contour value of the statistical bitmap, the gray average of the statistical bitmap, the gray variance of the statistical bitmap, and the function of the gray histogram of the statistical bitmap .
The method according to claim 1, wherein the stitching the compressed bitmap into the second image comprises: writing size information of each bitmap in the header file of the second image, in the Location information in a picture, compression coding algorithm information used.
A picture compression device, comprising:

Obtaining a module, configured to acquire a first picture;

a cutting module, configured to cut the first picture into a plurality of bitmaps;

a determining module, configured to determine a compression encoding algorithm for each bitmap in the plurality of bitmaps;

a compression module, configured to compress a corresponding bitmap by using a determined compression coding algorithm;

a splicing module, configured to splicing the compressed plurality of bitmaps into a second picture.
The apparatus according to claim 9, wherein the determining module comprises:

a determining unit, configured to determine a pixel quality of each bitmap in the plurality of bitmaps;

And a selecting unit, configured to select a compression coding algorithm for each bitmap according to the pixel quality.
The apparatus according to claim 10, wherein said selecting unit comprises:

a determining subunit, configured to determine whether the pixel quality of each bitmap in the plurality of bitmaps is less than a predetermined threshold;

Selecting a subunit for selecting the compression encoding algorithm of the bitmap in the first compression encoding algorithm set if the determination result is yes; otherwise, selecting the bitmap in the second compression encoding algorithm set The compression coding algorithm.