TWI772108B - Computational display and operating method thereof - Google Patents

Abstract

A computational display and an operating method thereof are disclosed. The computational display includes an image dividing module, a decomposition calculating module, a block combining module and a display module. The image dividing module is used to divide an initial light-field image into a plurality of target blocks. The decomposition calculating module is coupled to the image dividing module and used to independently apply a decomposition algorithm to each of the plurality of target blocks. The block combining module is coupled to the decomposition calculating module and used to combine the plurality of target blocks into a result light-field image. The display module including multi-layer display units is coupled to the block combining module and used to display result light-field image.

Description

Computational display and method of operation

The present invention relates to displays, and more particularly, to a computing display and a method of operation thereof.

In recent years, with the evolution of display technology, computational displays such as Multi-layer LCD Factorized Displays can expand applications such as wider dynamic range and wider color gamut, especially factorized light fields Displays have attracted widespread attention because they are significantly superior to traditional light field displays in terms of space, viewing angle resolution, display brightness, and composition volume.

Different from the traditional light field display, before the multi-layer liquid crystal decomposed light field display displays the light field image, the light field image needs to be decomposed into the content played on each layer of liquid crystal of the multi-layer liquid crystal decomposed light field display.

However, since portable electronic products have been widely used in the daily life of the public, if the multi-layer liquid crystal decomposed light field display is to be applied to portable electronic products, it will be faced with the problem of the memory of the portable electronic products. Limitations (such as the limited capacity of the built-in memory (On-chip memory) cannot meet the demand, and the external memory (Off-chip memory) is limited by the access bandwidth), resulting in poor efficiency of light field image decomposition, which seriously affects The display performance of the multi-layer liquid crystal split light field display needs to be overcome urgently.

In view of this, the present invention provides a computing display and an operation method thereof to effectively solve the above-mentioned problems encountered in the prior art.

One embodiment according to the present invention is a computing display. In this embodiment, the computing display includes an image segmentation module, a decomposition calculation module, a block combination module and a display module. The image segmentation module is used for segmenting the initial light field image into a plurality of target blocks. The decomposition algorithm module is coupled to the image segmentation module, and is used for independently applying the decomposition algorithm to each target block of the plurality of target blocks. The block combination module is coupled to the decomposition calculation module, and is used for combining the plurality of target blocks into a resultant light field image. The display module includes a multi-layer display unit and is coupled to the block combining module for displaying the resulting light field image.

In one embodiment, the computing display is a multi-layer LCD factorized display and at least one display unit of the multi-layer display units is a liquid crystal layer.

In one embodiment, the decomposition algorithm adopted by the decomposition calculation module is a block decomposition algorithm, and the decomposition calculation module includes: an asymmetric update unit, used to limit the update range in the plurality of target blocks to The block-based decomposition algorithm is implemented in an asymmetric update method, so that the decomposition results of multiple target blocks are free of decomposition defects.

In one embodiment, the asymmetric update unit divides one target block of the plurality of target blocks into an update region (Update Region) and a reference region (Reference Region), and executes the block-based method on the target block. When decomposing the algorithm, only the update region is updated and not the reference region.

In one embodiment, the update regions of any two adjacent target blocks in the plurality of target blocks do not overlap with each other and the reference regions thereof overlap with each other.

In one embodiment, the decomposition algorithm adopted by the decomposition calculation module is a block decomposition algorithm, and the decomposition calculation module includes: an external iterative (Inter-Block Iteration) unit for performing external processing on the plurality of target blocks. Iterating, so that any two adjacent target blocks in the plurality of target blocks can still maintain the same boundary after multiple iterations.

In one embodiment, the external iterative unit repeatedly tours each target block to perform external iteration, and the external iteration includes the respective internal iteration (Intra-Block Iteration) of each target block, so that the target block updated first can be The boundaries of the two are kept consistent with reference to the updated adjacent target blocks.

In one embodiment, the external iterative unit implements external execution of the plurality of target blocks by loading and storing between the outer memory and the block memory. iterate.

In one embodiment, the capacity of the external memory is greater than the capacity of the block memory, the target blocks divided by the image segmentation module are stored in the external memory, and the decomposition calculation module is used to store the target blocks. Blocks are loaded from external memory into block memory for decomposition calculations.

In one embodiment, the decomposition algorithm module is suitable for an embedded system with limited memory capacity.

Another embodiment according to the present invention is a method for operating a computing display. In this embodiment, the computing display operation method is used to operate the computing display. The operation method of the computational display comprises the following steps: (a) dividing an initial light field image into a plurality of target blocks; (b) applying a decomposition algorithm to each target block of the plurality of target blocks independently; ( c) combining the plurality of target blocks into a resulting light field image; and (d) displaying the resulting light field image.

In one embodiment, the computing display is a multi-layer liquid crystal decomposition display and at least one display unit of the multi-layer display units is a liquid crystal layer.

In one embodiment, the decomposition algorithm is a block decomposition algorithm, and the operation method of the computational display further includes: narrowing the update range of the plurality of target blocks, and executing the block decomposition algorithm in an asymmetric update manner method, so that the decomposition results of the multiple target blocks are free of decomposition defects.

In one embodiment, the computing display operation method further includes: dividing one target block of the plurality of target blocks into an update area and a reference area; and performing a block-based decomposition algorithm on the target block only Update the update area without updating the reference area.

In one embodiment, the update regions of any two adjacent target blocks in the plurality of target blocks do not overlap with each other and the reference regions thereof overlap with each other.

In one embodiment, the decomposition algorithm is a block decomposition algorithm, and the operation method of the computational display includes: performing external iteration on the plurality of target blocks, so that any two adjacent target areas in the plurality of target blocks are formed. Blocks remain bounded consistent across multiple iterations.

In one embodiment, the computational display operation method repeats iterating through each target block for external iteration, and the external iteration includes the respective internal iteration of each target block, so that the target block updated first can refer to the updated target block. The completed adjacent target blocks maintain the same boundary between the two.

In one embodiment, the computing display operating method implements external iteration on the plurality of target blocks through loading and storing between external memory and block memory.

In one embodiment, the capacity of the external memory is greater than the capacity of the block memory, the target blocks divided in step (a) are stored in the external memory, and the target blocks are stored in the external memory before step (b). A target block is loaded from external memory to block memory for decomposition calculation.

In one embodiment, the decomposition algorithm is suitable for embedded systems with limited memory capacity.

Compared with the prior art, the computing display and its operation method of the present invention use a block-based decomposition algorithm to process the light field image, so as to effectively solve the problem of poor light field image decomposition efficiency due to the memory limitation of portable electronic products Therefore, the display performance of the multi-layer liquid crystal split light field display can be improved and it can be applied to embedded systems with limited memory capacity. In addition, in order to avoid the generation of the block-based decomposition algorithm, the computing display and the operation method thereof of the present invention further propose a block definition method and a block-based decomposition update method to avoid the block-based decomposition algorithm from generating boundaries at the boundaries Decomposes the Blocky effect of flaws and boundary inconsistencies.

The advantages and spirit of the present invention can be further understood from the following detailed description of the invention and the accompanying drawings.

Reference will now be made in detail to the exemplary embodiments of the present invention, examples of which are illustrated in the accompanying drawings. Elements/components using the same or similar numbers in the drawings and the embodiments are intended to represent the same or similar parts.

One embodiment according to the present invention is a computing display. In this embodiment, the computing display may be any kind of factorized display, such as a multi-layer LCD factorized display, and the multi-layer display unit may include at least one liquid crystal layer and may include a backlight layer, But not limited to this.

Please refer to FIG. 1 , which is a functional block diagram of the computing display in this embodiment. As shown in FIG. 1 , the computing display 1 includes an image segmentation module 10 , a block loading module 12 , a decomposition calculation module 14 , a block storage module 16 , a block combination module 18 and a display module 19 . The image splitting module 10 is coupled to the block loading module 12 . The block loading module 12 is coupled to the decomposition calculation module 14 . The decomposition calculation module 14 is coupled to the block storage module 16 . The block storage module 16 is coupled to the block combination module 18 . The block combining module 18 is coupled to the display module 19 .

When the image segmentation module 10 receives the light field data LFD and the initial multi-frame multi-layer decomposed image M0, the image segmentation module 10 is used for dividing the initial multi-frame multi-layer decomposed image M0 into a plurality of target blocks according to the light field data LFD. It can be stored in memory. In detail, each layer of decomposed images in the initial multi-frame multi-layer decomposed images M0 is divided into a plurality of decomposed image blocks, and the corresponding decomposed image blocks in each layer of decomposed images together form a light field image block .

Next, the block loading module 12 loads the plurality of target blocks stored in the memory into the decomposition calculation module 14, and the decomposition calculation module 14 applies the block decomposition algorithm to the plurality of independent calculation modules. On each of the target blocks, after the decomposition calculation is completed, the decomposition results of the plurality of target blocks are stored in the block storage module 16 . The block combining module 18 is used for combining the decomposition results of the target blocks into a multi-frame multi-layer decomposition image M1. The display module 19 includes multi-layer display units (at least one layer of the display units is a liquid crystal layer) for displaying the resultant multi-frame multi-layer decomposed images M1.

As shown in FIG. 2 and FIG. 3 , the computing platform CP used in the present invention includes a block memory BM and a processor PR. The processor PR and the block memory BM are coupled to each other. The computing platform CP is coupled to the external memory BUS through the bus bar BUS. In this embodiment, the external memory BUS can be used to store the frame contents of a plurality of frames and the block memory BM of the computing platform CP can be used to store the block contents of the plurality of target blocks, which is processed by the processor PR. Multiple target blocks are iteratively multiplied to update.

In fact, the computing platform CPF used in the present invention may include a Field Programmable Gate Array (FPGA) circuit or an Application Specific Integrated Circuit (ASIC) to achieve the effect of accelerating computing , but not limited to this.

It should be noted that the difference between the block-based decomposition algorithm used in the present invention and the traditional decomposition algorithm is that the initial multi-frame multi-layer light field image is cut into a plurality of target blocks, and then the block-based decomposition algorithm is divided into multiple target blocks. The decomposition algorithm is independently applied to the plurality of target blocks, and finally the decomposition results of each target block are combined into a resultant multi-frame multi-layer light field image.

In practical applications, before each target block starts to operate, it can be loaded from the external memory OM to the block memory BM of the computing platform CP. Examples of the external memory OM and the block memory BM vary according to different computing platforms CP. On the memory level, the external memory OM usually has a larger capacity and can store complete light field images and decomposed images (ie, frame contents), but has a low access bandwidth. The block memory BM has a small capacity and can only store the content of one target block (that is, the block content), but it can be accessed by the processor PR that performs the operation more quickly, so it can effectively improve the block type. The execution efficiency of the decomposition algorithm.

After multiple iterative multiplication update operations are performed on the computing platform CP, the updated content of the target block in the block memory BM will be written back to the external memory OM via the bus BUS, and then the next target will be loaded. block and perform the corresponding operation, and the rest can be deduced by analogy. Under this computing architecture, it is not necessary to repeatedly read the data in the external memory OM during the iterative process, thereby effectively reducing the block-based decomposition algorithm's requirement for overall memory access, and also reducing the waiting time for memory access. time and additional power consumption caused by frequent access to external memory OM.

It should be noted that, since the direct application of the block-based decomposition algorithm may cause block effects such as the inconsistency between the decomposition defect of the target block and the boundary of the adjacent target blocks, it will appear as discontinuous image defects visually. , the present invention proposes the following methods to solve this problem.

First, the decomposition calculation module 14 of the computing display 1 in FIG. 1 may include an asymmetric update unit (not shown in the figure), which can be used to limit the update range in the plurality of target blocks, in an asymmetric update manner. Execute the block-based decomposition algorithm, so that the decomposition results of multiple target blocks are free of decomposition defects.

For example, as shown in FIG. 4 , the target block B0 in the display image DM can be defined as an update area UR and a reference area RR, and can be updated through asymmetrical updating (for example, only the update area UR is updated without updating the reference area RR) ) to execute the block-based decomposition algorithm, so as to effectively solve the decomposition defects at the boundary of the target block B0 due to the inconsistency with the definition of the traditional decomposition algorithm.

It should be noted that, in this embodiment, in the display image DM, the update regions UR in the two adjacent target blocks B0 and B1 do not overlap each other, but the reference regions RR of the two overlap each other, but Not limited to this. In another embodiment, the update regions UR in the two adjacent target blocks B0 and B1 may also overlap with each other.

Please also refer to FIG. 5 . FIG. 5 is a schematic diagram of Load and Store between the external memory OM and the block memory BM when an asymmetric update is performed. As shown in FIG. 5 , the process of performing asymmetric update for each target block of the display image DM may include the following steps (a) to (d):

(a) loading the target block (eg B0) of the display image DM into the block memory BM, wherein the target block includes the update area UR and the reference area RR;

(b) performing a block-based decomposition algorithm on the target block according to a preset number of iterative updates, wherein in the operation process, only the content of the update region UR of the target block will be updated, as for the reference region RR in the target block The content of the loaded part is not enough to represent the complete definition, so it will not be updated but only used as a reference for the operation;

(c) storing the updated content of the update area UR of the target block to the external memory OM; and

(d) Load the next target block (eg B1 ) into the block memory BM, and repeat the above steps (a) to (c).

It should be noted that, in the above-mentioned embodiments, a full load and a partial store can be adopted to effectively reduce the requirement of the block-based decomposition algorithm for overall memory access. The update sequence of the plurality of target blocks may be B0, B1, . . . as shown in FIG. 6, but not limited thereto.

The above-described embodiments can be implemented on memory-constrained embedded systems. Suppose a block-based decomposition accelerator for a 7x7x1920x1080 light field is implemented and the number of iterations is set to 100. Compared with the traditional decomposition algorithm, this embodiment can reduce the DRAM bandwidth requirement by 100 times, and can use the on-chip memory (On-chip-memory) with a bandwidth of only 69 kilobytes (Kbytes). This enables the block-based decomposition algorithm of the present invention to run on an embedded system with limited memory capacity.

In addition, the implementation of the block-based decomposition algorithm using the asymmetric update method in this embodiment can effectively eliminate image defects at the boundary of the target block. Compared with the traditional symmetrical update method, the image quality of the test image is improved. Peak Signal-to-Noise Ratio (PSNR) can be improved by about 10 decibels (dB) on average.

Based on the above-mentioned embodiment, after the original image is divided into a plurality of target blocks, no matter in which order the update is performed, the target block updated later will refer to the target block that has been updated. For example, in the block update sequence shown in FIG. 6 , the later updated block B1 refers to the updated part of the block B0 . On the contrary, the target block updated first will refer to the random starting content that has not been updated. Therefore, the block decomposition algorithm may still encounter inconsistencies at the boundaries, resulting in visual block effects.

In order to avoid the above problems, the present invention proposes an external iterative update method to keep the boundaries of the target blocks consistent. For example, the decomposition calculation module 14 of the computational display 1 in FIG. 1 may further include an external iterative (Inter-Block Iteration) unit (not shown) for performing external iteration on the plurality of target blocks, resulting in Any two adjacent target blocks in the plurality of target blocks can still maintain the same boundary after multiple iterations.

Please refer to FIG. 7 . FIG. 7 is a schematic diagram of repeatedly iterating through each target block to perform external iteration. As shown in FIG. 7 , the present invention can divide the iterative update of the overall operation into an inner iteration (Intra-block iteration) and an outer iteration, where “intra-block iteration” means that the processor PR pair on the computing platform CP is loaded into the region A single target block of the block memory BM performs the iterative update of the block-based decomposition algorithm, and "outer iteration" means to repeatedly tour each of the plurality of target blocks. In other words, an external iteration includes the respective internal iteration of each target block, which is also implemented through loads and stores between the external memory OM and the block memory BM. In this operation mode, in subsequent external iterations, the target block that is updated first will have the opportunity to refer to the adjacent target block that has been updated, so as to achieve an ideal update result, so that any two adjacent target blocks can be Block boundaries can remain consistent.

For example, the present invention can first preset the number of internal iterations and the number of external iterations according to the light field image reconstruction quality requirements, and initialize the decomposed image with random values. Next, the present invention can perform the following steps (e) to (h) until the number of external iterations is reached:

(e) load a target block from the external memory OM to the block memory BM on the computing platform CP according to the update sequence;

(f) performing the iterative update of the block decomposition algorithm for the target block until the end of the internal iteration;

(g) storing the updated content of the target block in the external memory OM; and

(h) Load the next target block and repeat steps (e) to (g) above until the end of the inner iteration of the last target block.

The above-described embodiments can be implemented on memory-constrained embedded systems. Suppose to implement a block-based decomposition accelerator for a 7x7x1920x1080 light field and set the number of iterations to 2 outer iterations and 50 inner iterations.

Compared with the traditional decomposition algorithm, the present embodiment can reduce the DRAM bandwidth requirement by 50 times, and can use the on-chip memory (On-chip-memory) with a bandwidth of only 69 kilobytes (Kbytes). This enables the block-based decomposition algorithm of the present invention to run on an embedded system with limited memory capacity.

Under the same calculation amount, although the DRAM bandwidth that can be saved in this embodiment is lower than that of the above-mentioned embodiment using the asymmetric update method due to the set number of external iterations, this embodiment further adopts the external iteration method to effectively eliminate the phase difference. The boundary inconsistency of adjacent target blocks. On the test image, compared with the embodiment using the asymmetric update method with the same amount of calculation, the image quality obtained by this embodiment further adopts the external iteration, and the peak signal-to-noise ratio (PSNR) of the image quality can be improved by about 2 decibels (dB) on average. .

Another embodiment according to the present invention is a method for operating a computing display. In this embodiment, the operation method of the computing display can be applied to any decomposition type display, such as a multi-layer liquid crystal decomposition type display, and at least one display unit of the multi-layer display units is a liquid crystal layer, but not limited thereto.

Please refer to FIG. 8 . FIG. 8 is a flowchart illustrating an operation method of the computing display in this embodiment. As shown in FIG. 8, the operation method of the computing display may include the following steps:

Step S10: dividing the initial light field image into multiple target blocks;

Step S12: independently applying the decomposition algorithm to each target block of the plurality of target blocks;

Step S14: combining the plurality of target blocks into a resulting light field image; and

Step S16: Display the resulting light field image.

In practical applications, the decomposition algorithm used in step S12 is a block decomposition algorithm, which can be applied to embedded systems with limited memory capacity, and the computing display operation method of the present invention can also limit the number of partitions. For the update range in each target block, the block-based decomposition algorithm is performed in an asymmetric update manner, so that the decomposition results of the plurality of target blocks have no block effect of decomposition defects, but not limited thereto.

It should be noted that, in order to avoid the block effect with inconsistent borders caused by the block decomposition algorithm, the computing display operation method of the present invention can divide one target block of the plurality of target blocks into an update area. and the reference area, and when the block-based decomposition algorithm is performed on the target block, only the update area is updated without updating the reference area, but not limited to this.

In one embodiment, the update regions of any two adjacent target blocks in the plurality of target blocks do not overlap with each other and the reference regions thereof overlap with each other. In another embodiment, the update regions of any two adjacent target blocks can also overlap each other, so there is no specific limitation.

In addition, the computing display operation method of the present invention can also perform external iteration on the plurality of target blocks, so that any two adjacent target blocks in the plurality of target blocks can still maintain the same boundary after multiple iterations , to avoid the block effect of inconsistent boundaries.

For example, the computing display operation method of the present invention can repeatedly traverse each target block to perform external iteration, and the external iteration includes the respective internal iteration of each target block, so that the target block updated first can refer to the previously updated target block. The updated adjacent target blocks maintain the same boundary between the two.

In one embodiment, the computing display operation method of the present invention can perform external iteration on the plurality of target blocks through loading and storing between the external memory and the block memory of the computing platform. Since the capacity of the external memory is greater than the capacity of the block memory, the target blocks segmented by performing step S10 in the computing display operating method of the present invention can be stored in the external memory. Before performing step S12, First, load the plurality of target blocks from the external memory to the block memory of the computing platform to perform decomposition calculation through a block decomposition algorithm.

Compared with the prior art, the computing display and its operation method of the present invention use a block-based decomposition algorithm to process the light field image, so as to effectively solve the problem of poor light field image decomposition efficiency due to the memory limitation of portable electronic products Therefore, the display performance of the multi-layer liquid crystal split light field display can be improved and it can be applied to embedded systems with limited memory capacity. In addition, in order to avoid the generation of the block-based decomposition algorithm, the computing display and the operation method thereof of the present invention further propose a block definition method and a block-based decomposition update method to avoid the block-based decomposition algorithm from generating boundaries at the boundaries Decomposes the blocking effect of imperfections and inconsistent boundaries.

1: Computing Display 10: Image segmentation module 12: Chunk Loading Module 14: Decomposition calculation module 16:Block storage module 18: Block combination module 19: Display module LFD: Light Field Data M0: Initial multi-frame multi-layer decomposed image M1: Resulting multi-frame multi-layer decomposed image OM: External memory CP: Computing Platform BM: block memory BUS: bus bar PR: Processor DM: Display image LFM: Light Field Image UR: update area RR: Reference area B0: block B1: Block S10~S16: Steps

The accompanying drawings of the present invention are described as follows: FIG. 1 is a functional block diagram of a computing display according to a preferred embodiment of the present invention. FIG. 2 is a schematic diagram illustrating the coupling between an external memory and a computing platform through a bus. FIG. 3 is a schematic diagram illustrating that the external memory stores the frame content and the block memory of the computing platform stores the block content and is iteratively multiplied and updated by the processor. FIG. 4 is a schematic diagram illustrating that only the update area is updated without updating the reference area when the block-based decomposition algorithm is executed. FIG. 5 is a schematic diagram of loading and storing between external memory and block memory during asymmetric update. FIG. 6 is a schematic diagram illustrating an update sequence of blocks. FIG. 7 is a schematic diagram of repeatedly iterating through each target block for external iteration. FIG. 8 is a flowchart illustrating an operation method of a computing display according to another preferred embodiment of the present invention.

1: Computing Display

10: Image segmentation module

12: Chunk Loading Module

14: Decomposition calculation module

16:Block storage module

18: Block combination module

19: Display module

LFD: Light Field Data

M0: Initial multi-frame multi-layer decomposed image

M1: Resulting multi-frame multi-layer decomposed image

Claims (20)

A computing display includes: an image dividing module for dividing an initial light field image into a plurality of target blocks; a decomposing arithmetic module, coupled to the image dividing module, for dividing a block-type image The decomposition algorithm is independently applied to each target block of the plurality of target blocks; a block combination module is coupled to the decomposition calculation module for combining the plurality of target blocks into a resultant light field an image; and a display module including a multi-layer display unit and coupled to the block combination module for displaying the resultant light field image. The computing display of claim 1, wherein the computing display is a multi-layer LCD factorized display and at least one display unit in the multi-layer display unit is a liquid crystal layer. The computing display as claimed in claim 1, wherein the decomposition calculation module comprises: an asymmetric update unit, used to limit the update range of the plurality of target blocks, and execute the block in an asymmetric update manner A decomposition algorithm is adopted, so that the decomposition results of the plurality of target blocks are free of decomposition defects. The computing display of claim 3, wherein the asymmetric update unit divides one target block of the plurality of target blocks into an update region (Update Region) and a reference region (Reference Region), and When executing the block-based decomposition algorithm on the target block, only the update area is updated without updating the reference area. The computing display of claim 4, wherein the update areas of any two adjacent target blocks in the plurality of target blocks do not overlap each other and the reference areas thereof overlap each other. The computing display of claim 1, wherein the decomposition calculation module comprises: an external iterative (Inter-Block Iteration) unit for performing external iteration on the plurality of target blocks, resulting in the plurality of target blocks Any two adjacent target blocks can still maintain the same boundary after many iterations. The computing display of claim 6, wherein the external iteration unit repeatedly loops each target block to perform the external iteration, and the external iteration includes a respective inner iteration (Intra-Block Iteration) of each target block. ), so that the target block updated first can refer to the adjacent target block that has been updated to maintain the same boundary between the two. The computing display of claim 6, wherein the external iteration unit is implemented by loading and storing between an external memory (Outer Memory) and a block memory (Block memory) The outer iteration is performed on the plurality of target blocks. The computing display of claim 8, wherein the capacity of the external memory is greater than the capacity of the block memory, and the target blocks divided by the image dividing module are stored in the external memory, The decomposition calculation module loads the plurality of target blocks from the external memory to the block memory to perform decomposition calculation. The computing display of claim 1, wherein the decomposition algorithm module is suitable for an embedded system (Embedded System) with limited memory capacity. A computing display operation method for operating a computing display, including Including the following steps: (a) dividing an initial light field image into a plurality of target blocks; (b) independently applying a block decomposition algorithm to each target block of the plurality of target blocks; ( c) combining the plurality of target blocks into a resultant light field image; and (d) displaying the resultant light field image. The method for operating a computing display as claimed in claim 11, wherein the computing display is a multi-layer liquid crystal decomposition display and at least one display unit in the multi-layer display unit is a liquid crystal layer. The computing display operation method of claim 11, wherein the computing display operation method further comprises: narrowing the update range of the target blocks, and executing the block-based decomposition algorithm in an asymmetric update manner , so that the decomposition results of the multiple target blocks have no decomposition defects. The computing display operation method of claim 13, further comprising: dividing one of the target blocks into an update area and a reference area; and executing the block on the target block When using the formula decomposition algorithm, only the update region is updated without updating the reference region. The computing display operation method of claim 14, wherein the update areas of any two adjacent target blocks in the plurality of target blocks do not overlap each other and the reference areas thereof overlap each other. The computing display operation method of claim 11, wherein the computing display operation method comprises: performing external iteration on the plurality of target blocks, so that any two adjacent target blocks in the plurality of target blocks are The bounds are still consistent after many iterations. The computational display operation method of claim 16, wherein the computational display operation method repeatedly iterates through each target block for the outer iteration, and the outer iteration includes the respective inner iteration of each target block, The target block that is updated first can refer to the adjacent target block that has been updated so that the boundaries of the two are consistent. The computing display operation method of claim 16, wherein the computing display operation method realizes the external execution of the plurality of target blocks through loading and storing between an external memory and a block memory iterate. The computing display operation method of claim 18, wherein the capacity of the external memory is greater than the capacity of the block memory, and the target blocks divided in step (a) are stored in the external memory , before step (b), load the plurality of target blocks from the external memory to the block memory for decomposition calculation. The computing display operation method of claim 11, wherein the block-based decomposition algorithm is suitable for an embedded system with limited memory capacity.

Images