Microscopic out-of-focus imaging deblurring method based on full-range context

Abstract

The invention relates to a microscopic out-of-focus imaging deblurring method based on a full-range context. The microscopic out-of-focus imaging deblurring method comprises the steps of constructing a microscopic imaging deblurring neural network model; obtaining a training data set, and training the microscopic imaging area fuzzy neural network model based on the training data set to obtain a trained microscopic imaging area fuzzy neural network model; and inputting a to-be-processed blurred image into the trained microscopic imaging area blurred neural network model to obtain a clear image. According to the method, from the full-range context information, on the basis of comprehensively considering factors such as background stray light, defocusing degree and detail texture of the microscopic image, high-quality reconstruction is carried out on the defocusing degraded image, the definition and identifiability of the microscopic image are effectively improved, and a powerful guarantee is provided for subsequent accurate observation and quantitative analysis.

Description

Technical Field

[0001] This application relates to the field of microscopic imaging. Specifically, it relates to a method for deblurring microscopic out-of-focus imaging based on full-range context. Background Art

[0002] With the continuous growth of the demand for microscopic observation, microscopic imaging has become an important means to obtain microscopic information in many fields such as biomedicine and materials science. However, microscopic systems are often affected by various factors, resulting in out-of-focus problems, which significantly reduce the clarity of the captured images, thereby limiting the accuracy and efficiency of observation and analysis. Once microscopic out-of-focus occurs, not only are the structural details in the image difficult to identify, but it may also cause deviations in subsequent quantitative analysis results, having an adverse impact on high-precision experiments and clinical diagnoses.

[0003] At the same time, with the rapid development of precision manufacturing, nano-detection, and high-resolution microscopic observation, the demand for high-quality and high-clarity microscopic images is constantly increasing. However, the microscopic imaging environment is usually complex, and factors such as sample thickness, uneven refractive index, and the performance of the optical system may all cause imaging out-of-focus. Especially when observing multi-layered structures or dynamic living organisms, traditional focusing or stack shooting methods often struggle to balance efficiency and accuracy, resulting in varying degrees of blurring in the large number of images obtained. Since the targets of microscopic imaging are usually small in size and have fine structural details, if the out-of-focus blur cannot be effectively restored, key information may be missed, and even the accurate identification and judgment of the target may be interfered with.

[0004] Currently, although the research on deblurring microscopic images in the academic and industrial circles at home and abroad has achieved initial results, most of them are aimed at motion blur or imaging blur of ordinary optical systems, and there are relatively few methods applicable to the microscopic out-of-focus scenario. It should be noted that there are significant differences in the causes between microscopic out-of-focus and general motion blur or optical scattering phenomena. The former is mainly caused by factors such as optical path mismatch and sample offset, and its result is manifested as overall or local defocus degradation; while the latter is often caused by reasons such as rapid movement of the object and medium scattering, and the imaging characteristics are different from those of microscopic out-of-focus. Therefore, directly applying existing deblurring algorithms to the microscopic out-of-focus scenario often fails to achieve ideal results. In addition, some deblurring techniques for microscopic images also have certain limitations, such as insufficient consideration of complex interferences such as imaging noise and background stray light, or lack of targeted modeling of the aberrations and off-axis imaging characteristics of the microscopic system, resulting in limited applicability in high-resolution and high-precision imaging scenarios. Summary of the Invention

[0005] In order to overcome at least one deficiency in the prior art, this application provides a method for deblurring microscopic out-of-focus imaging based on full-range context.

[0006] In a first aspect, a method for deblurring microscopic out-of-focus imaging based on full-range context is provided, including:

[0007] Construct a microscopic imaging deblurring neural network model; the microscopic imaging deblurring neural network model includes an encoder and a decoder. The encoder is used to gradually extract the deep features of the input blurred image, and the decoder is used to gradually restore the deep features back to the image space to obtain a clear image; the encoder includes a first convolutional layer and multiple encoding layers, and the decoder includes multiple decoding layers and a second convolutional layer. The features output by each encoding layer are input into the corresponding decoding layer through skip connections; the structures of the encoding layer and the decoding layer are the same, and both are full-range context blocks; the full-range context block is used to capture local features, regional features, and global features in the space;

[0008] Obtain a training data set, and train the microscopic imaging area deblurring neural network model based on the training data set to obtain a trained microscopic imaging area deblurring neural network model; the samples in the training data set include blurred images and corresponding real clear images; input the blurred image into the microscopic imaging area deblurring neural network model to obtain a predicted clear image; calculate the loss based on the predicted clear image and the real clear image to train the model;

[0009] Input the blurred image to be processed into the trained microscopic imaging area deblurring neural network model to obtain a clear image.

[0010] In one embodiment, the encoder includes 5 encoding layers, namely the first encoding layer, the second encoding layer, the third encoding layer, the fourth encoding layer, and the fifth encoding layer;

[0011] The decoder includes 5 decoding layers, namely the first decoding layer, the second decoding layer, the third decoding layer, the fourth decoding layer, and the fifth decoding layer;

[0012] The blurred image is input into the first convolutional layer to be converted into spatial features, and the spatial features sequentially pass through the first encoding layer, the second encoding layer, the third encoding layer, the fourth encoding layer, and the fifth encoding layer to obtain depth features;

[0013] The depth features are input into the first decoding layer, the first feature output by the first encoding layer is input into the fourth decoding layer, the second feature output by the second encoding layer is input into the third decoding layer, the third feature output by the third encoding layer is input into the second decoding layer, and the fourth feature output by the fourth decoding layer is input into the first decoding layer;

[0014] The features output by the fifth decoding layer pass through the second convolutional layer and are then connected in a residual manner with the blurred image to obtain a clear image.

[0015] In one embodiment, the full-range context block includes: a first local range block, a regional range block, a second local range block, and a global range block;

[0016] The input of the full - range context block is locally feature - enhanced through the first local - range block to obtain the first local feature; the first local feature is normalized and window - partitioned to obtain the first value vector and the first key vector;

[0017] The input of the full - range context block is normalized and window - partitioned to obtain the first query vector;

[0018] The first value vector, the first key vector, and the first query vector are input into the region - range block, and the region feature is obtained by using the multi - head self - attention mechanism;

[0019] The region feature, the first value vector, the first key vector, and the input of the full - range context block are added together to obtain the first addition result; the first addition result passes through the normalization layer and the MLP, and is added to the first addition result to obtain the second addition result;

[0020] The second addition result is input into the second local - range block for local feature enhancement to obtain the second local feature; the second local feature is normalized and window - partitioned to obtain the second value vector and the second key vector;

[0021] The second addition result is normalized and window - partitioned to obtain the partition result;

[0022] The second addition result passes through max - pooling, traditional attention operation, and the Sigmoid activation function, and is multiplied by the partition result to obtain the second query vector;

[0023] The second value vector, the second key vector, and the second query vector are input into the global - range block, and the global feature is obtained by using the multi - head self - attention mechanism;

[0024] The global feature, the second value vector, the second key vector, and the second addition result are added together to obtain the third addition result; the third addition result passes through the normalization layer and the MLP, and is added to the third addition result to obtain the output of the full - range context block.

[0025] In one embodiment, the structures of the first local - range block and the second local - range block are the same, including a depth - wise convolutional layer, a point - wise convolutional layer, a point - wise convolutional layer, and a depth - wise convolutional layer connected in sequence.

[0026] In a second aspect, a microscopic defocus imaging de - blurring device based on full - range context is provided, including:

[0027] A model construction module for constructing a microscopic imaging deblurring neural network model; the microscopic imaging deblurring neural network model includes an encoder and a decoder. The encoder is used to gradually extract the deep features of the input blurred image, and the decoder is used to gradually restore the deep features back to the image space to obtain a clear image. The encoder includes a first convolutional layer and multiple encoding layers, and the decoder includes multiple decoding layers and a second convolutional layer. The features output by each encoding layer are input into the corresponding decoding layer through skip connections. The structures of the encoding layer and the decoding layer are the same, and both are full-range context blocks. The full-range context block is used to capture local features, regional features, and global features in the space;

[0028] A training module for obtaining a training data set and training the microscopic imaging area deblurring neural network model based on the training data set to obtain a trained microscopic imaging area deblurring neural network model; the samples in the training data set include blurred images and corresponding real clear images. The blurred image is input into the microscopic imaging area deblurring neural network model to obtain a predicted clear image. The loss is calculated based on the predicted clear image and the real clear image to train the model;

[0029] A prediction module for inputting the blurred image to be processed into the trained microscopic imaging area deblurring neural network model to obtain a clear image.

[0030] In a third aspect, a computer-readable storage medium is provided. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements the above-mentioned microscopic defocus imaging deblurring method based on the full-range context.

[0031] In a fourth aspect, a computer program product is provided, including a computer program / instructions. When the computer program / instructions are executed by a processor, it implements the above-mentioned microscopic defocus imaging deblurring method based on the full-range context.

[0032] Compared with the prior art, the present application has the following beneficial effects: The microscopic defocus imaging deblurring method based on the full-range context of the present application starts from the full-range context information, and on the basis of comprehensively considering factors such as microscopic image background stray light, defocus degree, and detail texture, performs high-quality reconstruction of the defocus degraded image, effectively improving the clarity and recognizability of the microscopic image, and providing a strong guarantee for subsequent precise observation and quantitative analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The present application can be better understood by referring to the description given below in conjunction with the accompanying drawings. The drawings, together with the following detailed description, are included in this specification and form a part of this specification. In the drawings:

[0034] Figure 1 A schematic diagram of the microscopic imaging deblurring neural network model is shown;

[0035] Figure 2 A schematic diagram of the full - range context block is shown;

[0036] Figure 3 A schematic diagram of the local - range block is shown;

[0037] Figure 4 A structural block diagram of a microscopic defocus imaging de - blurring device based on the full - range context is shown. Detailed implementation manners

[0038] In the following, exemplary embodiments of the present application will be described in conjunction with the accompanying drawings. For the sake of clarity and conciseness, not all features of the actual embodiments are described in the specification. However, it should be understood that many embodiment - specific decisions may be made during the development of any such actual embodiment to achieve the specific goals of the developer, and these decisions may vary with different embodiments.

[0039] Here, it should also be noted that, in order to avoid obscuring the present application with unnecessary details, only the device structures closely related to the solution of the present application are shown in the drawings, and other details less related to the present application are omitted.

[0040] It should be understood that the present application is not limited to the described embodiments due to the following description with reference to the drawings. In this document, where feasible, embodiments can be combined with each other, features can be replaced or borrowed between different embodiments, and one or more features can be omitted in one embodiment.

[0041] The embodiments of the present application provide a microscopic defocus imaging de - blurring method based on the full - range context, which mainly includes the following steps:

[0042] Step S1, construct a microscopic imaging de - blurring neural network model; the microscopic imaging de - blurring neural network model includes an encoder and a decoder. The encoder is used to gradually extract the deep features of the input blurred image, and the decoder is used to gradually restore the deep features back to the image space to obtain a clear image; the encoder includes a first convolutional layer and multiple encoding layers, the decoder includes multiple decoding layers and a second convolutional layer, and the features output by each encoding layer are input into the corresponding decoding layer through skip connections; the structures of the encoding layer and the decoding layer are the same, and both are full - range context blocks; the full - range context block is used to capture local features, regional features, and global features in the space.

[0043] Step S2: Obtain a training data set and train a microscopic imaging region fuzzy neural network model based on the training data set to obtain a trained microscopic imaging region fuzzy neural network model. The samples in the training data set include blurred images and corresponding real clear images. Input the blurred images into the microscopic imaging region fuzzy neural network model to obtain predicted clear images (i.e., de-blurred images). Calculate the loss based on the predicted clear images and the real clear images to train the model.

[0044] Here, blurred images and corresponding real clear images are used as image pairs for the training and testing of the neural network. To meet the need for a large amount of data in deep learning and the different sizes of existing images, preprocess the input images. Generate multiple training samples from a single image by cropping, significantly expanding the data scale.

[0045] Step S3: Input the blurred image to be processed into the trained microscopic imaging region fuzzy neural network model to obtain a clear image.

[0046] In this embodiment, by introducing the full-range context modeling technology and integrating local, regional, and global features, an efficient and accurate de-blurring solution is provided for microscopic imaging, meeting the high-resolution imaging requirements in complex environments.

[0047] In one embodiment, Figure 1 shows a schematic diagram of the microscopic imaging de-blurring neural network model. Refer to Figure 1 The encoder includes 5 encoding layers, namely the first encoding layer, the second encoding layer, the third encoding layer, the fourth encoding layer, and the fifth encoding layer.

[0048] The decoder includes 5 decoding layers, namely the first decoding layer, the second decoding layer, the third decoding layer, the fourth decoding layer, and the fifth decoding layer.

[0049] The blurred image is input into the first convolutional layer to be converted into spatial features. The spatial features sequentially pass through the first encoding layer, the second encoding layer, the third encoding layer, the fourth encoding layer, and the fifth encoding layer to obtain depth features.

[0050] The depth features are input into the first decoding layer. The first feature output by the first encoding layer is input into the fourth decoding layer. The second feature output by the second encoding layer is input into the third decoding layer. The third feature output by the third encoding layer is input into the second decoding layer. The fourth feature output by the fourth decoding layer is input into the first decoding layer.

[0051] The features output by the fifth decoding layer pass through the second convolutional layer and are then residually connected to the blurred image to obtain a clear image.

[0052] In this embodiment, the microscopic imaging deblurring neural network model consists of a multi-layer symmetric encoder-decoder structure. The encoder is used to gradually extract the deep features of the input blurred microscopic image, and the decoder is used to gradually restore the features extracted by the encoder back to the image space, enhance and refine specific feature information, and obtain a high-quality clear image. The two perform feature fusion through skip connections, fusing detailed features and high-level semantic features, avoiding information loss, and at the same time enhancing the decoder's ability to restore local details, thereby improving the performance and deblurring effect of the overall network.

[0053] In one embodiment, Figure 2 The schematic diagram of the full-range context block is shown. The full-range context block includes: the first local range block, the regional range block, the second local range block, and the global range block;

[0054] The input of the full-range context block undergoes local feature enhancement through the first local range block to obtain the first local feature; the first local feature is normalized and windowed (using the normalization layer LN and the windowing layer WP) to obtain the first value vector V and the first key vector K; here, the input is divided into multiple non-overlapping windows, and shape adjustment operations are used to generate the value vector and the key vector. Here, the input to the full-range context block corresponding to the decoding layer is 2, and the 2 inputs are concatenated as the input to the full-range context block.

[0055] The input of the full-range context block is normalized and windowed (using the normalization layer LN and the windowing layer WP) to obtain the first query vector Q;

[0056] The first value vector, the first key vector, and the first query vector are input into the regional range block, and the multi-head self-attention mechanism (MSA) is used to obtain the regional features;

[0057] The regional features, the first value vector, the first key vector, and the input of the full-range context block are added together to obtain the first addition result; the first addition result passes through the normalization layer LN and the MLP (Multilayer Perceptron), and is added to the first addition result to obtain the second addition result, which is used to enhance the feature expression after attention, improve the expression effect and stability of the model.

[0058] The second addition result is input into the second local range block for local feature enhancement to obtain the second local feature; the second local feature is normalized and windowed (using the normalization layer LN and the windowing layer WP) to obtain the second value vector and the second key vector;

[0059] The second addition result is normalized and windowed (using the normalization layer LN and the windowing layer WP) to obtain the windowing result;

[0060] The second addition result passes through a max pooling Maxpool, a traditional attention operation Attention, and a Sigmoid activation function, and is multiplied by the partitioning result to obtain a second query vector;

[0061] The second value vector, the second key vector, and the second query vector are input into the global scope block, and the multi-head self-attention mechanism MSA is used to obtain global features;

[0062] The global features, the second value vector, the second key vector, and the second addition result are added together to obtain a third addition result; the third addition result passes through a normalization layer LN and an MLP, and is added to the third addition result to obtain the output of the full-scope context block.

[0063] Figure 3 The schematic diagram of the local scope block is shown. The structures of the first local scope block and the second local scope block are the same, including a depth convolution layer, a pointwise convolution layer, a pointwise convolution layer, and a depth convolution layer connected in sequence.

[0064] Here, first, the input features are spatially context-encoded using a 3×3 depth convolution, then the features are mapped to a low-dimensional space using a 1×1 pointwise convolution, then the features are non-linearly transformed using a Relu activation function, then, a 1×1 pointwise convolution is used to adjust the channel dimension to the number of input channels, and finally, a 3×3 depth convolution is used to further extract spatial features.

[0065] Specifically, the network model is trained by backpropagation of the L1 loss calculated based on the predicted clear image and the real clear image. The blurred microscopic image passes through the neural network and outputs the reconstructed clear image. The L1 loss is used to calculate the model gradient, enabling the network to continuously learn the mapping relationship between the blurred microscopic image and the clear image, constraining the training process, and updating the model parameters.

[0066] Adopting the same inventive concept as the microscopic defocus imaging deblurring method based on the full-scope context, this embodiment also provides a corresponding microscopic defocus imaging deblurring device based on the full-scope context, Figure 4 The structural block diagram of the microscopic defocus imaging deblurring device based on the full-scope context is shown, including:

[0067] A model construction module 41 for constructing a microscopic imaging deblurring neural network model. The microscopic imaging deblurring neural network model includes an encoder and a decoder. The encoder is used to gradually extract the deep features of the input blurred image, and the decoder is used to gradually restore the deep features back to the image space to obtain a clear image. The encoder includes a first convolutional layer and multiple encoding layers, and the decoder includes multiple decoding layers and a second convolutional layer. The features output by each encoding layer are input into the corresponding decoding layer through skip connections. The structures of the encoding layer and the decoding layer are the same, both being full-range context blocks. The full-range context block is used to capture local features, regional features, and global features in the space.

[0068] A training module 42 for obtaining a training data set and training the microscopic imaging area deblurring neural network model based on the training data set to obtain a trained microscopic imaging area deblurring neural network model. The samples in the training data set include blurred images and corresponding real clear images. The blurred images are input into the microscopic imaging area deblurring neural network model to obtain predicted clear images. The loss is calculated based on the predicted clear images and the real clear images to train the model.

[0069] A prediction module 43 for inputting the blurred image to be processed into the trained microscopic imaging area deblurring neural network model to obtain a clear image.

[0070] The microscopic defocus imaging deblurring device based on the full-range context in this embodiment has the same inventive concept as the above-mentioned microscopic defocus imaging deblurring method based on the full-range context. Therefore, the specific implementation of this device can be seen in the embodiment part of the above-mentioned microscopic defocus imaging deblurring method based on the full-range context, and its technical effects correspond to those of the above method, which will not be elaborated here.

[0071] This application embodiment provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it realizes the above-mentioned microscopic defocus imaging deblurring method based on the full-range context.

[0072] This application embodiment provides a computer program product, including a computer program / instructions. When the computer program / instructions are executed by a processor, it realizes the above-mentioned microscopic defocus imaging deblurring method based on the full-range context.

[0073] In summary, this application has the following technical effects: The microscopic defocus imaging deblurring method based on the full-range context of this application starts from the full-range context information. On the basis of comprehensively considering factors such as background stray light, defocus degree, and detailed texture of microscopic images, it performs high-quality reconstruction of defocused degraded images, effectively improving the clarity and recognizability of microscopic images, and providing a strong guarantee for subsequent precise observation and quantitative analysis.

[0074] As described above, these are only various embodiments of the present application. However, the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A microscopic defocus imaging deblurring method based on full-range context, characterized in that Including: Constructing a microscopic imaging deblurring neural network model; the microscopic imaging deblurring neural network model includes an encoder and a decoder. The encoder is used to gradually extract the deep features of the input blurred image, and the decoder is used to gradually restore the deep features back to the image space to obtain a clear image. The encoder includes a first convolutional layer and multiple encoding layers, and the decoder includes multiple decoding layers and a second convolutional layer. The features output by each encoding layer are input into the corresponding decoding layer through skip connections. The structures of the encoding layer and the decoding layer are the same, both being full-range context blocks. The full-range context block is used to capture local features, regional features, and global features in space; Obtaining a training data set, training the microscopic imaging area deblurring neural network model based on the training data set to obtain a trained microscopic imaging area deblurring neural network model; the samples in the training data set include blurred images and corresponding real clear images. Inputting the blurred image into the microscopic imaging area deblurring neural network model to obtain a predicted clear image. Calculating the loss based on the predicted clear image and the real clear image to train the model; Inputting the blurred image to be processed into the trained microscopic imaging area deblurring neural network model to obtain a clear image.

2. The method according to claim 1, characterized in that, The encoder includes 5 encoding layers, namely the first encoding layer, the second encoding layer, the third encoding layer, the fourth encoding layer, and the fifth encoding layer; The decoder includes 5 decoding layers, namely the first decoding layer, the second decoding layer, the third decoding layer, the fourth decoding layer, and the fifth decoding layer; The blurred image is input into the first convolutional layer to be transformed into spatial features, and the spatial features sequentially pass through the first encoding layer, the second encoding layer, the third encoding layer, the fourth encoding layer, and the fifth encoding layer to obtain depth features; The depth features are input into the first decoding layer, the first feature output by the first encoding layer is input into the fourth decoding layer, the second feature output by the second encoding layer is input into the third decoding layer, the third feature output by the third encoding layer is input into the second decoding layer, and the fourth feature output by the fourth decoding layer is input into the first decoding layer; The features output by the fifth decoding layer pass through the second convolutional layer and are connected in a residual manner with the blurred image to obtain a clear image.

3. The method according to claim 1, characterized in that The full-range context block includes: a first local range block, a regional range block, a second local range block, and a global range block; The input of the full-range context block undergoes local feature enhancement through the first local range block to obtain a first local feature. The first local feature is normalized and window-divided to obtain a first value vector and a first key vector; The input of the full-range context block is normalized and window-divided to obtain a first query vector; The first value vector, the first key vector, and the first query vector are input into the regional range block, and the multi-head self-attention mechanism is used to obtain regional features; The input of the region feature, the first value vector, the first key vector, and the full-range context block are added together to obtain a first addition result; the first addition result passes through a normalization layer and an MLP, and is added to the first addition result to obtain a second addition result; The second addition result is input into the second local range block for local feature enhancement to obtain a second local feature; the second local feature is normalized and windowed to obtain a second value vector and a second key vector; The second addition result is normalized and windowed to obtain a partitioning result; The second addition result passes through max pooling, a traditional attention operation, and a Sigmoid activation function, and is multiplied by the partitioning result to obtain a second query vector; The second value vector, the second key vector, and the second query vector are input into the global range block, and a multi-head self-attention mechanism is used to obtain global features; The global feature, the second value vector, the second key vector, and the second addition result are added together to obtain a third addition result; the third addition result passes through a normalization layer and an MLP, and is added to the third addition result to obtain the output of the full-range context block.

4. The method according to claim 3, wherein The structures of the first local range block and the second local range block are the same, including a depth convolution layer, a pointwise convolution layer, a pointwise convolution layer, and a depth convolution layer connected in sequence.

5. A microscopic defocus imaging deblurring device based on full-range context, characterized in that, Comprising: A model construction module for constructing a microscopic imaging deblurring neural network model; the microscopic imaging deblurring neural network model includes an encoder and a decoder. The encoder is used to gradually extract deep features of the input blurred image, and the decoder is used to gradually restore the deep features back to the image space to obtain a clear image. The encoder includes a first convolutional layer and multiple encoding layers, and the decoder includes multiple decoding layers and a second convolutional layer. The features output by each encoding layer are input into the corresponding decoding layer through skip connections. The structures of the encoding layer and the decoding layer are the same, both being full-range context blocks. The full-range context block is used to capture local features, regional features, and global features in space; A training module for obtaining a training data set and training the microscopic imaging region deblurring neural network model based on the training data set to obtain a trained microscopic imaging region deblurring neural network model. The samples in the training data set include blurred images and corresponding real clear images. The blurred images are input into the microscopic imaging region deblurring neural network model to obtain predicted clear images. The model is trained based on the loss calculated from the predicted clear images and the real clear images; A prediction module for inputting the blurred image to be processed into the trained microscopic imaging region deblurring neural network model to obtain a clear image.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which when executed by a processor, implements the full-range context-based microscopic defocus imaging deblurring method according to any one of claims 1-4.

7. A computer program product, characterized in that, Comprising a computer program / instructions, when the computer program / instructions are executed by a processor, to implement the method for defocus imaging deblurring based on the full-range context according to any one of claims 1-4.

Images