Image data processing method and device, equipment and medium

Abstract

The embodiment of the invention provides an image data processing method and device, equipment and a medium. The image data processing method comprises the following steps: reading compressed data and decoding the compressed data to obtain decoded wavelet transform data; wherein the wavelet transform data comprises a plurality of high-frequency coefficients and a plurality of low-frequency coefficients; performing at least one level of two-dimensional wavelet inverse transformation according to the wavelet transformation data to obtain compensation data; wherein the two-dimensional wavelet inverse transformation comprises the following steps: determining first even number sequences according to the low-frequency coefficients, and determining first odd number sequences according to the high-frequency coefficients and the two adjacent first even number sequences; determining a second even number sequence according to the first even number sequence; determining a second odd sequence according to the first odd sequence and the second even sequence adjacent to the first odd sequence; determining the compensation data according to the second even sequence and the second odd sequence; and carrying out image picture compensation on a picture to be displayed based on the compensation data.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to the technical field of image processing, and in particular to a processing method and device for image data, equipment and a medium. BACKGROUND

[0002] In the production process of an OLED (Organic Light-Emitting Diode) display screen, due to reasons such as material purity and process level, some products may have the phenomenon of uneven display brightness (Mura).

[0003] Eliminating Mura is a common technical problem in the display field. At present, an external compensation method is generally used to eliminate Mura. Optical data of the display screen is collected through an optical device, and compensation data is calculated according to the collected optical data.

[0004] However, since the SRAM (Static Random Access Memory) resource inside a display driving chip, such as an ODDI (OLED Display Driver IC), is very limited, in order to obtain better picture quality, data compression processing needs to be performed on the Demura compensation data. SUMMARY

[0005] Therefore, embodiments of the present application provide a processing method and device for image data, equipment and a medium.

[0006] In one aspect, a processing method for image data is provided, comprising:

[0007] reading compressed data and performing decoding to obtain wavelet transform data after decoding; wherein the wavelet transform data comprises a plurality of high-frequency coefficients and a plurality of low-frequency coefficients;

[0008] performing at least one level of two-dimensional wavelet inverse transform according to the wavelet transform data to obtain compensation data; wherein the two-dimensional wavelet inverse transform comprises:

[0009] determining a first even sequence according to the low-frequency coefficients, and determining a first odd sequence according to the high-frequency coefficients and two adjacent first even sequences;

[0010] determining a second even sequence according to the first even sequence, and determining a second odd sequence according to the first odd sequence and the second even sequence adjacent to the first odd sequence;

[0011] determining the compensation data according to the second even sequence and the second odd sequence;

[0012] based on the compensation data, performing image frame compensation on a to-be-displayed frame.

[0013] In another aspect, an embodiment of the present application provides a processing device for image data, comprising:

[0014] a decoding module configured to read compressed data and perform decoding to obtain decoded wavelet transform data; wherein the wavelet transform data comprises a plurality of high-frequency coefficients and a plurality of low-frequency coefficients;

[0015] a first transform module configured to perform at least one level of two-dimensional inverse wavelet transform based on the wavelet transform data to obtain compensation data; wherein the two-dimensional inverse wavelet transform comprises:

[0016] determining a first even sequence based on the low-frequency coefficients, and determining a first odd sequence based on the high-frequency coefficients and two adjacent first even sequences;

[0017] determining a second even sequence based on the first even sequence, and determining a second odd sequence based on the first odd sequence and the second even sequence adjacent to the first odd sequence;

[0018] determining the compensation data based on the second even sequence and the second odd sequence;

[0019] a compensation module configured to perform image frame compensation on a to-be-displayed frame based on the compensation data.

[0020] In still another aspect, an embodiment of the present application provides a processing method for image data, comprising:

[0021] obtaining compensation data of an image frame;

[0022] performing at least one level of two-dimensional wavelet transform on the compensation data to obtain wavelet transform data; the wavelet transform data comprises a plurality of high-frequency coefficients and a plurality of low-frequency coefficients; wherein the two-dimensional wavelet transform comprises:

[0023] dividing the compensation data into a second odd sequence and a second even sequence; determining a first even sequence based on the second even sequence, and determining a first odd sequence based on the second odd sequence and two adjacent second even sequences of the second odd sequence; determining the low-frequency coefficients based on the first even sequence, and determining the high-frequency coefficients based on the first odd sequence and two adjacent first even sequences of the first odd sequence;

[0024] performing encoding on the wavelet transform data to obtain compressed data.

[0025] In another aspect, an embodiment of the present application further provides an electronic device, comprising a memory, a processor, and a display screen; the memory stores a program; and the processor implements any of the above-mentioned processing methods when executing the program;

[0026] The display screen is used to: display the compensated picture to be displayed when the processor is configured to execute the processing method.

[0027] On the other hand, an embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, any of the above-mentioned processing methods is implemented.

[0028] In an embodiment of the present application, at least one level of a two-dimensional 1 / 3 wavelet inverse transform algorithm is performed on the wavelet transform data to obtain compensation data. This two-dimensional 1 / 3 wavelet inverse transform algorithm includes: determining a first even sequence based on low-frequency coefficients, and determining a first odd sequence based on high-frequency coefficients and two adjacent first even sequences; determining a second even sequence based on the first even sequence; and determining a second odd sequence based on the first odd sequence and two adjacent second even sequences; and determining compensation data based on the second even sequence and the second odd sequence. This simplifies the calculation process, facilitates RTL (Register Transfer Level) hardware implementation, and reduces SRAM resource consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 A flowchart of a method for processing image data provided in an embodiment of the present application;

[0030] Figure 2 A flowchart of an algorithm for a two-dimensional wavelet inverse transform provided in an embodiment of the present application;

[0031] Figure 3 A simplified diagram of the wavelet transform algorithm provided in an embodiment of the present application;

[0032] Figure 4 A schematic diagram of a two-dimensional wavelet transform provided in an embodiment of the present application;

[0033] Figure 5 A schematic diagram of a partitioned two-dimensional wavelet transform provided in an embodiment of the present application;

[0034] Figure 6 A schematic diagram of partitions with different compressed data lengths provided in an embodiment of the present application;

[0035] Figure 7 Schematic diagram of obtaining compensated data by performing two-dimensional wavelet inverse transform on compressed data in an embodiment of the present application;

[0036] Figure 8 FIG. 2 is a schematic diagram of a compensation data obtained by performing two-dimensional wavelet inverse transform on compressed data corresponding to a partition in an embodiment of the present application;

[0037] Figure 9 FIG. 4 is a flowchart of another image data processing method provided in an embodiment of the present application;

[0038] Figure 10 FIG. 6 is a flowchart of an algorithm of two-dimensional wavelet transform provided in an embodiment of the present application;

[0039] Figure 11 FIG. 8 is a structural schematic diagram of an image data processing device provided in an embodiment of the present application;

[0040] Figure 12 FIG. 10 is a structural schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0041] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the embodiments of the present application and the accompanying drawings. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of the present application.

[0042] In the following description, a large number of specific details are given to provide a more complete understanding of the present application. However, it is obvious to those skilled in the art that the present application can be implemented without one or more of these details. In other examples, in order to avoid obscuring the present application, some technical features known in the art are not described; that is, all features of the actual embodiments are not described in detail, and well-known functions and structures are not described in detail.

[0043] The terms used herein are only for the purpose of describing specific embodiments and not as a limitation of the present application. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the terms "comprise" and / or "comprising", when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. As used herein, the term "and / or" includes any and all combinations of the associated listed items.

[0044] For a thorough understanding of the present application, detailed steps and detailed structures will be presented in the following description in order to illustrate the technical solutions of the present application. The preferred embodiments of the present application are described in detail as follows, however, in addition to these detailed descriptions, the present application can also have other implementation manners.

[0045] The compression processing of Demura compensation data is a technology for reducing the storage space occupied by compensation data through a specific algorithm under the condition that the storage resources of a display driving chip are limited, while maintaining the correction accuracy to the maximum.

[0046] However, the current compression processing method has low compression and occupies more hardware resources, greatly limiting the applicability of OLED products.

[0047] Therefore, the embodiments of the present disclosure provide a processing method of image data, which involves data decompression, as shown in Figure 1 and Figure 2 The method comprises the following steps.

[0048] Step S101, reading compressed data and decoding to obtain wavelet transform data after decoding; wherein the wavelet transform data comprises a plurality of high frequency coefficients and a plurality of low frequency coefficients;

[0049] Step S102, performing at least one level of two-dimensional wavelet inverse transform according to the wavelet transform data to obtain compensation data; wherein the two-dimensional wavelet inverse transform comprises:

[0050] Step S21, determining a first even sequence according to the low frequency coefficients, and determining a first odd sequence according to the high frequency coefficients and the two first even sequences adjacent to each other;

[0051] Step S22, determining a second even sequence according to the first even sequence, and determining a second odd sequence according to the first odd sequence and the second even sequence adjacent to the first odd sequence;

[0052] Step S23, determining the compensation data according to the second even sequence and the second odd sequence;

[0053] Step S103, performing image picture compensation on a to-be-displayed picture based on the compensation data.

[0054] In the embodiments of the present disclosure, the compressed data comprises data generated after wavelet transform and data compression are performed on Demura compensation data. In some embodiments, the compressed data can be stored in a flash memory (Flash) outside the display driving chip. In other embodiments, the compressed data can be stored in a SRAM inside the display driving chip. In other embodiments, the compressed data can be transmitted in real time through a host interface.

[0055] In a specific embodiment, the compressed data is read and decoded to obtain the decoded wavelet transform data, including: reading the compressed data stored in the Flash outside the display driver chip, and buffering the compressed data in the SRAM inside the display driver chip. Then, the compressed data is decoded by using the special decoding engine inside the display driver chip to obtain the decoded data (i.e., the wavelet transform data). It should be noted that since the compressed data is obtained by wavelet transforming the image data and then performing data compression processing, the data obtained after decoding the compressed data is referred to as wavelet transform data.

[0056] In the embodiments of the present disclosure, the wavelet transform data includes a plurality of high-frequency coefficients and a plurality of low-frequency coefficients. Here, the high-frequency coefficients include abrupt change components extracted from the image data, and can be used to represent components of the non-uniformity of the abrupt change in the micro-scale space of the display screen. The low-frequency coefficients include smooth components extracted from the image data, and are used to represent components of the non-uniformity of the gradual change in the large-scale space of the display screen.

[0057] Wavelet transform is a new transform analysis method, which inherits and develops the localization idea of STFT (Short-Time Fourier Transform), and overcomes the shortcomings that the window size does not change with the frequency. It can provide a "time-frequency" window that changes with the frequency, and is an ideal tool for signal time-frequency analysis and processing.

[0058] The essence of wavelet transform is similar to Fourier transform, both of which use carefully selected bases to represent signal equations. Each wavelet transform has a wavelet function (mother wavelet) and a scale function (father wavelet). The base function of any wavelet transform is actually a set of scaling and translation of the mother wavelet and the father wavelet.

[0059] The wavelet transform algorithm in the embodiments of the present disclosure is a 1 / 3 wavelet transform algorithm improved from the 5 / 3 wavelet transform algorithm, which is a discrete wavelet transform (DWT) algorithm, and the implementation steps thereof include three steps: decomposition, prediction, and update. The decomposition is to divide the image data into two parts: an initial even sequence and an initial odd sequence. The prediction is to obtain a predicted odd sequence according to the initial even sequence obtained by decomposition, and to obtain a prediction error according to the predicted odd sequence and the initial odd sequence, and the prediction error is the high-frequency coefficient of the transform. The update is to set the initial even sequence as the low-frequency coefficient of the transform.

[0060] The 1 / 3 wavelet transform algorithm proposed in the embodiments of the present disclosure is as follows:

[0061]

[0062] Among them, x(2n+1) is the initial odd sequence, c(2n+1) is the high frequency coefficient, is a predicted odd sequence predicted based on the adjacent initial even sequence x(2n) and the initial even sequence x(2n+2). n is an integer greater than or equal to 0.

[0063] d(2n)=x(2n); (2)

[0064] Among them, x(2n) is the initial even-numbered sequence, and d(2n) is the low-frequency coefficient.

[0065] The 1 / 3 wavelet inverse transform algorithm can be obtained by inverting the above 1 / 3 wavelet transform algorithm:

[0066] x(2n)=d(2n); (3)

[0067] Here, d(2n) is a low-frequency coefficient, and x(2n) is a first even-numbered sequence (ie, an initial even-numbered sequence).

[0068]

[0069] Among them, x(2n+1) is the first odd number sequence (i.e., the initial odd number sequence), c(2n+1) is the high frequency coefficient, is a predicted odd sequence predicted based on the adjacent first even sequence x(2n) and the first even sequence x(2n+2).

[0070] Figure 3 shows a simplified diagram of the algorithm. A, B, C, D, E, F, G, and H are multiple initial sequences arranged in sequence. A, C, E, and G are initial even-numbered sequences, and B, D, F, and H are initial odd-numbered sequences. Averaging adjacent initial even-numbered sequences yields a predicted odd-numbered sequence. By calculating the difference between the initial odd-numbered sequence and the predicted odd-numbered sequence, we can obtain the high-frequency coefficients H0, H1, H2, and H3. Based on the initial even-numbered sequence, we can determine the low-frequency coefficients L0, L1, L2, and L3.

[0071] For example, the predicted odd sequence is obtained based on the adjacent initial even sequences A and C ( Figure 3 (not shown), then the difference between the initial odd sequence B and the predicted odd sequence is calculated to obtain the high-frequency coefficient H0. The low-frequency coefficient L0 can be determined based on the initial even sequence A.

[0072] Here, it should be noted that the resolution of OLED screens is generally an even number, so the number of rows and columns of Demura compensation data is also an even number. Therefore, the embodiment of the present disclosure only considers the case where the initial sequence rows and columns are an even number. Figure 3As shown, in order to obtain the last high frequency coefficient, symmetric extension is needed to be made to the right edge of the initial sequence. Specifically, another initial even sequence G is extended to the right of the initial odd sequence H. After wavelet transform, the initial sequence can be divided into two parts, i.e. low frequency coefficients and high frequency coefficients, and the number of the low frequency coefficients is equal to that of the high frequency coefficients and is half of the number of the initial sequence.

[0073] In some embodiments, the above-mentioned discrete wavelet transform is repeated on the low frequency coefficients to obtain the wavelet transform result of the next layer.

[0074] In the embodiments of the present disclosure, since the initial sequence is arranged on a two-dimensional plane, two-level wavelet transform is needed in the process of data processing on the initial sequence to obtain the wavelet transform data, that is, two-dimensional wavelet transform is needed.

[0075] Specifically, one-dimensional wavelet transform is made from the horizontal and vertical directions respectively to realize two-dimensional wavelet transform. As shown in the following formula (3) and formula (4). Figure 4 In the formula, a represents the original image, b represents the result of one-level wavelet transform, h1 represents the high frequency coefficient in the horizontal direction, v1 represents the high frequency coefficient in the vertical direction, c1 represents the high frequency coefficient in the diagonal direction, b1 represents the low frequency coefficient of 2 down-sampling, and c represents the continuous wavelet transform.

[0076] The two-dimensional wavelet transform algorithm in the embodiments of the present disclosure is realized based on the one-dimensional 1 / 3 wavelet transform algorithm, so the two-dimensional wavelet transform algorithm is a two-dimensional 1 / 3 wavelet transform algorithm. And since the two-dimensional wavelet inverse transform is the inverse of the two-dimensional wavelet transform, the two-dimensional wavelet inverse transform algorithm is a two-dimensional 1 / 3 wavelet inverse transform algorithm.

[0077] In combination with the above formula (3) and formula (4), the two-dimensional wavelet inverse transform in the embodiments of the present disclosure specifically includes: determining a first even sequence x(2n) according to the low frequency coefficient d(2n), and determining a first odd sequence x(2n+1) according to the high frequency coefficient c(2n+1) and the adjacent first even sequences x(2n) and x(2n+2); determining a second even sequence x(2n) according to the first even sequence x(2n), and determining a second odd sequence x(2n+1) according to the first odd sequence x(2n+1) and the adjacent second even sequences x(2n) and x(2n+2) of the first odd sequence x(2n+1); and determining compensation data according to the second even sequence x(2n) and the second odd sequence x(2n+1). ’ ’ ’ ’ ’ ’

[0078] ​​​​​​It should be noted that the "adjacent" mentioned in the embodiments of the present disclosure refers to the adjacent on the sequence number, for example, the sequences corresponding to the sequence numbers n and (n+1) are adjacent sequences.

[0079] In the embodiments of the present disclosure, at least one level of two-dimensional wavelet inverse transform can be performed on the wavelet transform data to obtain compensation data. Here, the compensation data includes a set of mathematical parameters for correcting the Mura defect of the display screen by the user, and the essence is to apply accurate voltage / current adjustment to each pixel in the display screen through the driving circuit to offset the physical deviation in the manufacturing process of the display screen.

[0080] In the embodiments of the present disclosure, after the two-dimensional 1 / 3 wavelet transform, the high-frequency coefficients and the low-frequency coefficients of the compensation data are separated. Since most of the high-frequency coefficients of the compensation data are small, there is a correlation between different levels of high-frequency coefficients. Therefore, data compression can be performed on the data after the two-dimensional 1 / 3 wavelet transform to obtain a higher compression ratio. That is, less SRAM resources can be used to store the compressed data. At the same time, the two-dimensional wavelet transform algorithm in the embodiments of the present disclosure can simplify the calculation process and facilitate the RTL hardware implementation. It can be understood that based on the advantages of the two-dimensional wavelet transform algorithm in the embodiments of the present disclosure, the two-dimensional wavelet inverse transform algorithm in the embodiments of the present disclosure can also reduce the use of SRAM resources and simplify the calculation process.

[0081] In some embodiments, the compressed data is read and decoded to obtain the decoded wavelet transform data, including:

[0082] reading the compressed data;

[0083] The compressed data is decoded based on a Set Partitioning In Hierarchical Trees (SPIHT) algorithm to obtain the decoded wavelet transform data.

[0084] Here, the SPIHT algorithm is an efficient image compression and encoding technology based on wavelet transform. The main idea of the SPIHT algorithm is to compare each coefficient in the set of wavelet coefficients with a given threshold value, and if it is greater than the threshold value, output a binary number as an indicator of the importance of the coefficient; this binary flag is the code stream after the compensation data is encoded. After all the wavelet transform coefficient values are traversed, the threshold value is halved, and the above process is repeated.

[0085] The SPIHT algorithm is improved based on the EZW (Embedded Zerotree Wavelets Encoding) algorithm, and is based on the zero-tree structure of the EZW algorithm. The tree nodes and all subsequent nodes are planned as a same set, three table items of LSP (List of Significant Pixels), LIP (List of Insignificant Pixels) and LIS (List of Insignificant Sets) are created, and a proper set partitioning and sorting strategy is adopted. The embedded coding is completed through four sub-processes of initialization, sorting process, refinement process and quantization step updating.

[0086] In some embodiments, the compressed data is read and decoded to obtain decoded wavelet transform data, including:

[0087] The compressed compensation data in one partition is read and decoded to obtain sub-data corresponding to the partition.

[0088] Specifically, the compressed data includes compressed compensation data located in multiple partitions. When the compressed data is read, only the compressed compensation data located in one partition can be read. In the decoding stage, only the compressed compensation data of the read partition can be decoded to obtain sub-data corresponding to the partition.

[0089] It can be understood that, since the compressed data includes compressed compensation data located in multiple partitions, the decoded compressed data can obtain sub-data corresponding to multiple partitions. In the embodiments of the present disclosure, the data obtained after decoding the compressed data is also referred to as wavelet transform data, that is, the wavelet transform data can include sub-data of multiple partitions.

[0090] In some embodiments, the wavelet transform data is subjected to two-dimensional wavelet inverse transform to obtain compensation data, including:

[0091] The two-dimensional wavelet inverse transform is performed on the sub-data of each partition to obtain compensation data corresponding to each partition.

[0092] In the embodiments of the present disclosure, the compressed data corresponding to different partitions can be sequentially subjected to data processing to obtain compensation data in a preset order. Each time the compressed data of one partition is subjected to data processing, the process includes: reading the compressed data corresponding to the partition and decompressing to obtain sub-data corresponding to the partition; performing two-dimensional wavelet inverse transform on the sub-data to obtain compensation data corresponding to the partition.

[0093] It should be noted that the traditional two-dimensional wavelet transform is generally a multi-level wavelet transform on the whole image. The embodiments of the present disclosure mainly aim at the data processing of Demura compensation data, such as data compression, decompression and two-dimensional wavelet inverse transform of Demura compensation data. These data processing of Demura compensation data need to be implemented inside the IC (for example, a display driving chip). It can be understood that if the traditional two-dimensional wavelet transform is used to perform multi-level wavelet transform on the whole image, the real-time performance of data processing will be poor and a large amount of SRAM resources will be consumed.

[0094] The method of partition processing is adopted in the embodiments of the present disclosure to solve the problems of real-time performance and SRAM resources. Since the wavelet transform is performed by down-sampling by 2, the number of levels of wavelet transform depends on whether the image resolution can be divided by 2 times. The compensation data is divided into multiple partitions, such as partition 1 to partition 5, and then two-dimensional wavelet transform is performed on each partition respectively. Since there is no correlation between the partitions, the data processing in the partition can be completed without waiting for the data of the rear partition of the image. Therefore, the real-time performance of the algorithm can be improved, and a large amount of SRAM resources is not needed to cache too much data, and a better improvement effect of image quality can be achieved. Figure 5

[0095] In some embodiments, determining the first even sequence according to the low-frequency coefficients comprises determining a plurality of low-frequency coefficients d(2n) as the first even sequence x(2n), as shown in formula (3).

[0096] In some embodiments, determining the second even sequence according to the first even sequence comprises determining the first even sequence x(2n) as the second even sequence x ’ (2n).

[0097] In some embodiments, after determining the compensation data corresponding to the partition according to the second even sequence and the second odd sequence, the compressed compensation data in the next partition is read and decoded to obtain the sub-data corresponding to the next partition.

[0098] It should be noted that in some embodiments, after obtaining the compensation data of the previous partition, the compressed compensation data in the next partition can be directly read and decoded. In this way, the operation of reading the compressed compensation data in the next partition and decoding the compressed compensation data can be synchronized with the operation of performing image frame compensation on the compensation data corresponding to the previous partition, so that the operation time can be saved. In other embodiments, after obtaining the compensation data of the previous partition, image frame compensation can be performed based on the compensation data of the partition, and then the compressed compensation data in the next partition is read and decoded.

[0099] ​In some embodiments, based on the compensation data, image frame compensation is performed on the to-be-displayed picture, including:

[0100] Based on the compensation data of one partition, image frame compensation is performed on the corresponding partition in the to-be-displayed picture to obtain to-be-displayed partition data;

[0101] The to-be-displayed partition data is cached.

[0102] In the embodiments of the present disclosure, the compensation data of one partition includes a plurality of compensation data values corresponding to a plurality of pixels in the partition. The image frame compensation performed on the corresponding partition in the to-be-displayed picture includes: in the corresponding partition in the to-be-displayed picture, summing the current image data value (for example, the gray scale offset value) of each pixel in the partition and the compensation data value corresponding to the pixel to realize image frame compensation and obtain to-be-displayed partition data.

[0103] In the embodiments of the present disclosure, the to-be-displayed partition data can be cached in a buffer register (Buffer).

[0104] In some embodiments, there can be no Mura in some partitions, that is, the partition has no corresponding compensation data, in which case the current image data of the partition can be directly cached.

[0105] In some embodiments, the image data processing method further includes:

[0106] In response to having cached all to-be-displayed partition data of the to-be-displayed picture, the compensated to-be-displayed picture is output.

[0107] In the embodiments of the present disclosure, when the compensation data of all partitions is obtained, and based on all the compensation data, image frame compensation is performed on all the partitions in the to-be-displayed picture to obtain all the to-be-displayed partition data, the compensated to-be-displayed picture can be output. At this time, compared with the original image, the Mura in the to-be-displayed picture is improved.

[0108] Next, a specific example of the image data processing method is described:

[0109] Step 1: Read the compressed data of one partition. Specifically, in order to ensure the real-time performance of the algorithm and reduce the SRAM resource, the whole picture needs to be partitioned for processing, as shown in FIG. 1. Figure 5 Since the data of the whole partition is compressed together in the wavelet transform-based compression algorithm, the decompression process also needs to be completed after the complete decompression of the data of the same partition. Here, the compressed data of each partition in the Flash is independent of each other. Assuming that there are 5 partitions in the Flash, the compressed data of the 5 partitions is as shown in FIG. 2. Figure 6It is to be noted that the compression ratio of the SPIHT compression algorithm is not fixed, and thus the compressed data length of each partition can be different. Thus, the data of different partitions can be distinguished in the decompression process.

[0110] Step 2: SPIHT decoding, specifically: since the SPIHT decoding is performed in layers, with the decrease of the threshold, the precision of the decompressed data is higher and higher. Thus, the wavelet transform data of the partition can be obtained only after the lowest threshold is used to run completely. It is to be noted that the data in the intermediate process can be cached in the SRAM. The partition compression reduces the SRAM cost of this part.

[0111] Step 3: two-dimensional wavelet inverse transform, specifically: since the SPIHT algorithm compresses the data after the wavelet transform, the compensation data can be obtained only after the wavelet inverse transform is calculated again, as shown in Figure 7 .

[0112] Here, the two-dimensional wavelet inverse transform calculation is completed in two steps: 1. the data of the horizontal wavelet transform is obtained by the vertical wavelet inverse transform according to formulas (3) and (4); and 2. the final result is obtained by the horizontal wavelet inverse transform according to formulas (3) and (4), as shown in Figure 8 .

[0113] Step 4: data compensation, specifically: the compensation data generally stores the gray scale offset value calculated according to the optical data, and the decompressed compensation data can be directly added to the gray scale value at the corresponding position to realize the data compensation.

[0114] Step 5: reading the compressed data of the next partition, specifically: after the data of the current partition (for example, partition 1) is processed, the decompressed data in the SRAM can be put into the output buffer (for example, the buffer), the data in the current SRAM is emptied, and the decompression process of the data of the next partition (for example, partition 2) is entered.

[0115] As shown in Figure 9 and Figure 10 , the embodiment of the disclosure further provides a processing method of image data, which involves data compression, and the method comprises the following steps:

[0116] Step S201: acquiring compensation data of an image frame;

[0117] Step S202: performing at least one level of two-dimensional wavelet transform on the compensation data to obtain wavelet transform data; the wavelet transform data comprises a plurality of high-frequency coefficients and a plurality of low-frequency coefficients; wherein the two-dimensional wavelet transform comprises:

[0118] Step S31, dividing the compensation data into a second odd sequence and a second even sequence; determining a first even sequence according to the second even sequence, and determining a first odd sequence according to the second odd sequence and two second even sequences adjacent to the second odd sequence;

[0119] Step S31, determining a low-frequency coefficient according to the first even sequence, and determining a high-frequency coefficient according to the first odd sequence and two first even sequences adjacent to the first odd sequence;

[0120] Step S203, encoding the wavelet transform data to obtain compressed data.

[0121] In the embodiments of the present disclosure, the compensation data of the image frame includes: obtaining compensation data corresponding to any partition in the image frame.

[0122] In the embodiments of the present disclosure, the two-dimensional wavelet transform is specifically 1 / 3 two-dimensional wavelet transform, which can be understood based on the above-mentioned 1 / 3 two-dimensional wavelet inverse transform. Specifically, in combination with the above-mentioned formula (1) and formula (2), the two-dimensional wavelet transform includes: dividing the compensation data into a second odd sequence x ’ (2n+1) and a second even sequence x ’ (2n); determining a first even sequence x ’ (2n) according to the second even sequence x ’ (2n+1) and adjacent second even sequences x ’ (2n) and x ’ (2n+2); determining a first odd sequence x

[0123] In the embodiments of the present disclosure, after the two-dimensional 1 / 3 wavelet transform, the high-frequency coefficients and the low-frequency coefficients of the compensation data are separated. Since most of the high-frequency coefficients of the compensation data are small, there is a correlation between different levels of high-frequency coefficients. Therefore, data compression can obtain a higher compression ratio for the data after the two-dimensional 1 / 3 wavelet transform. At the same time, the two-dimensional wavelet transform algorithm in the embodiments of the present disclosure can simplify the calculation process and facilitate RTL hardware implementation.

[0124] In some embodiments, determining the first even sequence according to the second even sequence includes: determining the second even sequence x ’ (2n) as the first even sequence x

[0125] Determining the low-frequency coefficient according to the first even sequence includes: determining the first even sequence x (2n) as the low-frequency coefficient d (2n), as shown in formula (2).

[0126] In some embodiments, the compensation data is subjected to at least one level of two-dimensional wavelet transform to obtain wavelet transform data, including:

[0127] The compensation data is divided into a plurality of partitions;

[0128] The compensation data of each partition is subjected to at least one level of two-dimensional wavelet transform to obtain sub-data corresponding to the partition.

[0129] Here, it can be understood that, compared with the scheme of subjecting the partition compensation data to one level of two-dimensional wavelet transform, the scheme of subjecting the compensation data of each partition to two or more levels of two-dimensional wavelet transform can further improve the compression ratio, and less SRAM is required.

[0130] In some embodiments, the wavelet transform data is encoded to obtain compressed data, including:

[0131] Each sub-data of each partition is encoded to obtain compressed data corresponding to the partition.

[0132] In some embodiments, the wavelet transform data is encoded to obtain compressed data, including:

[0133] The wavelet transform data is encoded based on the SPIHT algorithm to obtain compressed data.

[0134] In the embodiments of the present disclosure, the specific content of the image data processing method related to data compression can be understood based on the image data processing method related to data decompression, which will not be repeated here.

[0135] The present disclosure also provides an image data processing apparatus 100, as shown in Figure 11 The image data processing apparatus 100 can be used to decompress data, and the image data processing apparatus 100 includes:

[0136] The decoding module 110 is configured to read and decode the compressed data to obtain wavelet transform data after decoding; wherein the wavelet transform data includes a plurality of high-frequency coefficients and a plurality of low-frequency coefficients;

[0137] The first transform module 120 is configured to perform at least one level of two-dimensional wavelet inverse transform on the wavelet transform data to obtain compensation data; wherein the two-dimensional wavelet inverse transform includes:

[0138] A first even sequence is determined according to the low-frequency coefficients, and a first odd sequence is determined according to the high-frequency coefficients and the two adjacent first even sequences;

[0139] A second even sequence is determined according to the first even sequence, and a second odd sequence is determined according to the first odd sequence and the second even sequence adjacent to the first odd sequence;

[0140] determining compensation data according to the second even sequence and the second odd sequence;

[0141] The compensation module 130 is configured to perform image compensation on the to-be-displayed picture based on the compensation data.

[0142] In some embodiments, the compressed data is stored in a storage module located outside the image data processing apparatus 100. In some examples, the storage module comprises a non-volatile memory.

[0143] In the embodiments of the present disclosure, the first transform module 120 comprises a discrete wavelet inverse transform module.

[0144] In the embodiments of the present disclosure, the image data processing apparatus 100 for decompressing data comprises an integrated circuit (IC). The image data processing apparatus 100 can read the compressed data in the storage module into a pre-allocated SRAM in a start-up stage. Specifically, the decompression process of the image data processing apparatus 100 comprises: the decoding module 110 first reads the compressed data of one partition from the storage module; then, the decoding module 110 decodes to obtain the two-dimensional wavelet transform data of the partition by applying the SPIHT algorithm; then, the first transform module 120 obtains the compensation data by two-dimensional wavelet inverse transform and compensates the image data by using the compensation module 130 to eliminate Mura. It can be understood that the above process can be repeated to complete the compensation of the entire image. After performing image compensation on the to-be-displayed picture, the compensated image data can be output.

[0145] In some embodiments, the wavelet transform data comprises a plurality of partition sub-data; the first transform module 120 is further configured to perform two-dimensional wavelet inverse transform on each partition sub-data respectively to obtain the compensation data corresponding to each partition.

[0146] In some embodiments, the decoding module 110 is further configured to read and decode the compressed compensation data in one partition to obtain the sub-data corresponding to the partition; and

[0147] After determining the compensation data corresponding to the partition according to the second even sequence and the second odd sequence, the compressed compensation data in the next partition is read and decoded to obtain the sub-data corresponding to the next partition.

[0148] In some embodiments, the compensation module 130 is further configured to:

[0149] perform image compensation on the corresponding partition in the to-be-displayed picture based on the compensation data of the partition to obtain to-be-displayed partition data;

[0150] cache the to-be-displayed partition data.

[0151] In some embodiments, the image data processing apparatus 100 further comprises:

[0152] The output module 140 is configured to output the compensated to-be-displayed picture in response to having buffered all to-be-displayed partition data of the to-be-displayed picture.

[0153] As shown in Figure 12 The embodiments of the present disclosure further provide an electronic device 300, comprising a memory 310, a processor 320 and a display screen 330; the memory 310 stores a program; the processor 320 implements any of the above processing methods when executing the program.

[0154] The display screen is configured to display the compensated to-be-displayed picture in the case that the processor 320 is configured to execute the processing method.

[0155] The embodiments of the present disclosure further provide a computer readable storage medium, which stores a computer program, and the computer program is executed by the processor to implement any of the above processing methods.

[0156] The above electronic device and computer readable storage medium belong to the same inventive concept as the above image data processing apparatus provided by the embodiments of the present disclosure, and the specific implementation and scheme effects can refer to the detailed description in the above embodiments, which will not be repeated here.

[0157] Those skilled in the art can understand that all or part of the steps of the above method embodiments can be completed by a program instruction related to hardware, and the foregoing program can be stored in a computer readable storage medium, and the program is executed to perform the steps of the above method embodiments; and the foregoing storage medium includes mobile storage devices, read only memory (ROM), magnetic discs or optical discs and various media that can store program codes.

[0158] Alternatively, the above integrated units of the present application, if realized in the form of software functional modules and sold or used as independent products, can also be stored in a computer readable storage medium. The storage medium includes mobile storage devices, ROM, magnetic discs or optical discs and various media that can store program codes.

[0159] The various embodiments / implementation manners provided by the present application can be combined with each other without contradiction.

[0160] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for processing image data, characterized in that: include: Reading the compressed data and decoding it to obtain decoded wavelet transform data; wherein the wavelet transform data includes: a plurality of high-frequency coefficients and a plurality of low-frequency coefficients; Performing at least one level of two-dimensional wavelet inverse transform on the wavelet transform data to obtain compensation data; wherein the two-dimensional wavelet inverse transform includes: Determine a first even-numbered sequence according to the low-frequency coefficients, and determine a first odd-numbered sequence according to the high-frequency coefficients and two adjacent first even-numbered sequences; Determine a second even-numbered sequence based on the first even-numbered sequence; and determine a second odd-numbered sequence based on the first odd-numbered sequence and the second even-numbered sequence adjacent to the first odd-numbered sequence; determining the compensation data according to the second even-numbered sequence and the second odd-numbered sequence; Based on the compensation data, image compensation is performed on the image to be displayed.

2. The processing method according to claim 1, characterized in that Determining the first even-numbered sequence according to the low-frequency coefficients includes: determining the plurality of low-frequency coefficients as the first even-numbered sequence; Determining a second even-numbered sequence according to the first even-numbered sequence includes: determining the first even-numbered sequence as the second even-numbered sequence.

3. The processing method according to claim 1, characterized in that The wavelet transform data includes a plurality of partitioned sub-data; The step of performing a two-dimensional wavelet inverse transform on the wavelet transform data to obtain compensation data includes: The two-dimensional wavelet inverse transform is performed on the sub-data of each partition respectively to obtain compensation data corresponding to each partition.

4. The processing method according to any one of claims 1 to 3, characterized in that: The method of reading compressed data and decoding to obtain decoded wavelet transform data includes: Reading the compressed compensation data in a partition and decoding it to obtain sub-data corresponding to the partition; and After determining the compensation data corresponding to the partition according to the second even sequence and the second odd sequence, the compressed compensation data in the next partition is read and decoded to obtain the sub-data corresponding to the next partition.

5. The processing method according to claim 3, characterized in that: The performing image compensation on the image to be displayed based on the compensation data includes: Based on the compensation data of a partition, performing image compensation on the corresponding partition in the picture to be displayed to obtain data of the partition to be displayed; caching the partition data to be displayed; The method further comprises: In response to all the to-be-displayed partition data of the to-be-displayed picture being buffered, the compensated to-be-displayed picture is output.

6. An image data processing device, characterized in that: include: A decoding module is configured to read the compressed data and decode it to obtain decoded wavelet transform data; wherein the wavelet transform data includes: a plurality of high-frequency coefficients and a plurality of low-frequency coefficients; The first transformation module is configured to perform at least one level of two-dimensional wavelet inverse transformation on the wavelet transformation data to obtain compensation data; wherein the two-dimensional wavelet inverse transformation includes: Determine a first even-numbered sequence according to the low-frequency coefficients, and determine a first odd-numbered sequence according to the high-frequency coefficients and two adjacent first even-numbered sequences; Determine a second even-numbered sequence based on the first even-numbered sequence; and determine a second odd-numbered sequence based on the first odd-numbered sequence and the second even-numbered sequence adjacent to the first odd-numbered sequence; determining the compensation data according to the second even-numbered sequence and the second odd-numbered sequence; The compensation module is configured to perform image compensation on the image to be displayed based on the compensation data.

7. The processing device according to claim 6, characterized in that The wavelet transform data includes sub-data of a plurality of partitions; the first transform module is further configured to perform the two-dimensional wavelet inverse transform according to the sub-data of each partition respectively, to obtain compensation data corresponding to each partition.

8. The processing device according to claim 6 or 7, characterized in that The decoding module is further configured to read the compensation data compressed in a partition and decode it to obtain sub-data corresponding to the partition; as well as After determining the compensation data corresponding to the partition according to the second even sequence and the second odd sequence, the compressed compensation data in the next partition is read and decoded to obtain the sub-data corresponding to the next partition.

9. The processing device according to claim 7, characterized in that The compensation module is further configured to: Based on the compensation data of a partition, performing image compensation on the corresponding partition in the picture to be displayed to obtain data of the partition to be displayed; caching the partition data to be displayed; The processing device further includes: The output module is configured to output the compensated picture to be displayed in response to all the buffered partition data to be displayed of the picture to be displayed.

10. A method for processing image data, characterized in that: include: Acquiring compensation data of the image frame; Performing at least one level of two-dimensional wavelet transform on the compensation data to obtain wavelet transform data; The wavelet transform data includes a plurality of high-frequency coefficients and a plurality of low-frequency coefficients; wherein the two-dimensional wavelet transform includes: Dividing the compensated data into a second odd sequence and a second even sequence; determining a first even sequence based on the second even sequence, and determining a first odd sequence based on the second odd sequence and two second even sequences adjacent to the second odd sequence; determining the low-frequency coefficients based on the first even sequence, and determining the high-frequency coefficients based on the first odd sequence and two adjacent first even sequences; The wavelet transformed data is encoded to obtain compressed data.

11. An electronic device, characterized in that: The device comprises a memory, a processor and a display screen; the memory stores a program; the processor implements the processing method according to any one of claims 1 to 5 or 10 when executing the program; The display screen is used to: display the compensated picture to be displayed when the processor is configured to execute the processing method according to any one of claims 1 to 5.

12. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processing method according to any one of claims 1 to 5 or 10 is implemented.

Images