Image processing method and device and sediment content measuring and calculating method

Abstract

The invention relates to an image processing method and device and a sediment content measuring and calculating method, and the image processing method comprises the steps: carrying out the spectrum collection of a target region through employing a same quantum dot spectrum sensor, obtaining a plurality of first spectrum images, and extracting a plurality of second spectrum images according to a preset image extraction rule; and performing image extraction on the plurality of first spectral images to obtain an image set corresponding to each spectral channel, performing average processing on all the extracted images in each image set to obtain an average image corresponding to each spectral channel, and determining a second spectral image based on the average image corresponding to all the spectral channels. According to the image processing method and device and the sediment content measuring and calculating method disclosed by the embodiment of the invention, image ghosting in the spectral image can be effectively eliminated, meanwhile, high measurement precision is kept, and the spectral image of the high-speed moving object is accurately obtained by utilizing the quantum dot spectral sensor so as to accurately calculate the sediment content.

Description

Technical Field

[0001] The present disclosure relates to the field of image processing technology, and in particular to an image processing method and device, and a sediment content measurement method. Background Art

[0002] In the field of measuring high-speed moving objects, image sensors are core components, and their performance directly determines the accuracy and reliability of the measurement results. However, in existing solutions, image sensor-based measurement systems often face a key problem when capturing images of high-speed moving objects: obvious afterimages will appear in the captured images. This afterimage is caused by the rapid change in position of the moving object in a very short time, and the exposure time and frame rate of the image sensor cannot fully match this high-speed change, resulting in the superposition of motion states at multiple time points in the image, thus forming blur or ghosting. This problem also occurs when acquiring spectral images. Therefore, there is an urgent need for a technical solution that can effectively eliminate the image afterimages generated when image sensors measure high-speed moving objects while maintaining high measurement accuracy. Summary of the invention

[0003] In view of this, the present disclosure proposes an image processing method and device, and a sediment content measurement method, which can effectively eliminate image afterimages in spectral images while maintaining high measurement accuracy, and realize the accurate acquisition of spectral images of high-speed moving objects using quantum dot spectral sensors.

[0004] According to one aspect of the present disclosure, there is provided an image processing method, comprising: using the same quantum dot spectral sensor to perform spectral acquisition on a target area to obtain a plurality of first spectral images, wherein the quantum dot spectral sensor is provided with a plurality of quantum dot regions, each of the quantum dot regions has optical characteristics corresponding to a different spectral channel, and each of the first spectral images includes spectral information under all spectral channels; performing image extraction on the plurality of first spectral images respectively according to a preset image extraction rule to obtain an image set corresponding to each of the spectral channels, each of the image sets including a plurality of extracted images of the corresponding spectral channels, all extracted images in the same image set come from different first spectral images, and each of the extracted images in the same image set has the same position in the corresponding first spectral image; performing averaging processing on all the extracted images in each of the image sets to obtain an average image corresponding to each of the spectral channels, and determining a second spectral image based on the average image corresponding to all of the spectral channels, the second spectral image including spectral information under all of the spectral channels after eliminating measurement errors.

[0005] In a possible implementation, according to a preset image extraction rule, image extraction is performed on the multiple first spectral images respectively to obtain an image set corresponding to each spectral channel, including: using the image in the target full-spectrum extraction area in each of the first spectral images as the first image corresponding to each first spectral image, and the target full-spectrum extraction area is a continuous area covering all spectral channels; obtaining multiple first sub-images by segmenting each of the first images based on the position information of each of the quantum dot areas; determining the image set corresponding to each of the spectral channels based on all the first sub-images, and the spectral channels corresponding to each first sub-image of the same first image are different; or using the image in each target spectral channel extraction area in each of the first spectral images as the multiple second sub-images corresponding to each first spectral image; and determining the image set corresponding to each of the spectral channels based on all the second sub-images.

[0006] In a possible implementation, the method further includes: determining the clarity of each of the first spectral images; when the number of first spectral images whose clarity is greater than or equal to a clarity threshold is zero, determining a preset default full-spectrum extraction area as the target full-spectrum extraction area; when the number of first spectral images whose clarity is greater than or equal to a clarity threshold is not zero, determining a selected full-spectrum extraction area determined based on the first spectral image with the highest clarity as the target full-spectrum extraction area.

[0007] In a possible implementation manner, the method further includes: determining each of the first spectral images and / or each of the extracted images as an image to be corrected, and performing image correction on each of the images to be corrected.

[0008] In a possible implementation, the quantum dot spectral sensor is further provided with a reference area, in which no quantum dots are deployed or the concentration of the deployed quantum dots is less than a preset concentration threshold; the image correction includes at least one of the following: when the spatial distribution of light intensity corresponding to the spectral acquisition satisfies a preset uniform distribution condition, a global Gaussian distribution is determined based on the reference area, and the global Gaussian distribution is used to correct the first signal corresponding to the quantum dot area to obtain a corrected image; when the spatial distribution of light intensity satisfies a preset non-uniform distribution condition, interpolation processing is performed based on the first signal corresponding to the quantum dot area and the second signal corresponding to the reference area, or high-order correction is performed based on the quantum dot area and the reference area to obtain a corrected image.

[0009] In a possible implementation, the method further includes: performing anomaly detection on pixel values ​​at all positions in each of the extracted images to obtain a detection result for each of the extracted images, wherein the detection result indicates the number of abnormal pixel values ​​in the extracted image, and the anomaly detection is performed based on a preset pixel interval; based on the detection result of each of the extracted images, deleting or correcting the abnormal extracted image.

[0010] In a possible implementation, all the extracted images in each of the image sets are averaged to obtain an average image corresponding to each of the spectral channels, including: for each of the image sets, the average value of each pixel position is calculated based on the pixel values ​​of all the extracted images in the image set at the same pixel position, and the average image of the corresponding spectral channel is determined according to the average value of all pixel positions or the average value adjusted based on the corresponding pixel reference value; and / or, a second spectral image is determined based on the average image corresponding to all spectral channels, including: according to a predetermined order of the spectral channels, the average images of the corresponding spectral channels are spliced ​​to obtain the second spectral image.

[0011] In a possible implementation manner, the method further includes: constructing a target mapping relationship representing an error between real spectral information and measured spectral information based on the real spectral image corresponding to the target area and the second spectral image.

[0012] According to another aspect of the present disclosure, an image processing device is provided, including: a spectrum acquisition module, used to use the same quantum dot spectrum sensor to perform spectrum acquisition on a target area to obtain multiple first spectrum images, wherein the quantum dot spectrum sensor is provided with multiple quantum dot regions, each of the quantum dot regions has optical characteristics corresponding to different spectral channels, and each of the first spectrum images includes spectrum information under all spectral channels; an image extraction module, used to perform image extraction on the multiple first spectrum images respectively according to a preset image extraction rule to obtain an image set corresponding to each of the spectral channels, each of the image sets includes multiple extracted images of the corresponding spectral channels, all extracted images in the same image set come from different first spectrum images, and each of the extracted images in the same image set has the same position in the corresponding first spectrum image; an average processing module, used to perform average processing on all the extracted images in each of the image sets to obtain an average image corresponding to each of the spectral channels, and determine a second spectrum image based on the average image corresponding to all the spectral channels, and the second spectrum image includes spectrum information under all spectral channels after eliminating measurement errors.

[0013] In a possible implementation, according to a preset image extraction rule, image extraction is performed on the multiple first spectral images respectively to obtain an image set corresponding to each spectral channel, including: using the image in the target full-spectrum extraction area in each of the first spectral images as the first image corresponding to each first spectral image, and the target full-spectrum extraction area is a continuous area covering all spectral channels; obtaining multiple first sub-images by segmenting each of the first images based on the position information of each of the quantum dot areas; determining the image set corresponding to each of the spectral channels based on all the first sub-images, and the spectral channels corresponding to each first sub-image of the same first image are different; or using the image in each target spectral channel extraction area in each of the first spectral images as the multiple second sub-images corresponding to each first spectral image; and determining the image set corresponding to each of the spectral channels based on all the second sub-images.

[0014] In a possible implementation, the device also includes a determination module, which is used to: determine the clarity of each of the first spectral images; when the number of first spectral images with clarity greater than or equal to a clarity threshold is zero, determine a preset default full-spectrum extraction area as the target full-spectrum extraction area; when the number of first spectral images with clarity greater than or equal to a clarity threshold is not zero, determine a selected full-spectrum extraction area determined based on the first spectral image with the highest clarity as the target full-spectrum extraction area.

[0015] In a possible implementation manner, the device further includes a correction module, which is used to: determine each of the first spectral images and / or each of the extracted images as an image to be corrected, and perform image correction on each of the images to be corrected.

[0016] In a possible implementation, the quantum dot spectral sensor is further provided with a reference area, in which no quantum dots are deployed or the concentration of the deployed quantum dots is less than a preset concentration threshold; the image correction includes at least one of the following: when the spatial distribution of light intensity corresponding to the spectral acquisition satisfies a preset uniform distribution condition, a global Gaussian distribution is determined based on the reference area, and the global Gaussian distribution is used to correct the first signal corresponding to the quantum dot area to obtain a corrected image; when the spatial distribution of light intensity satisfies a preset non-uniform distribution condition, interpolation processing is performed based on the first signal corresponding to the quantum dot area and the second signal corresponding to the reference area, or high-order correction is performed based on the quantum dot area and the reference area to obtain a corrected image.

[0017] In a possible implementation, the device also includes an anomaly detection module, which is used to: perform anomaly detection on pixel values ​​at all positions in each of the extracted images to obtain a detection result of each of the extracted images, wherein the detection result represents the number of abnormal pixel values ​​in the extracted image, and the anomaly detection is performed based on a preset pixel interval; based on the detection result of each of the extracted images, delete or correct the abnormal extracted image.

[0018] In a possible implementation, all the extracted images in each of the image sets are averaged to obtain an average image corresponding to each of the spectral channels, including: for each of the image sets, the average value of each pixel position is calculated based on the pixel values ​​of all the extracted images in the image set at the same pixel position, and the average image of the corresponding spectral channel is determined according to the average value of all pixel positions or the average value adjusted based on the corresponding pixel reference value; and / or, a second spectral image is determined based on the average image corresponding to all spectral channels, including: according to a predetermined order of the spectral channels, the average images of the corresponding spectral channels are spliced ​​to obtain the second spectral image.

[0019] In a possible implementation manner, the device further includes a construction module, which is used to construct a target mapping relationship representing an error between real spectral information and measured spectral information based on the real spectral image corresponding to the target area and the second spectral image.

[0020] According to another aspect of the present disclosure, a method for measuring sediment content is provided, comprising: performing calculation based on a second spectral image determined by using the above-mentioned image processing method to obtain the sediment content in a target area corresponding to the second spectral image.

[0021] According to another aspect of the present disclosure, an electronic device is provided, comprising: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to implement the above-mentioned image processing method when executing the instructions stored in the memory.

[0022] According to another aspect of the present disclosure, a non-volatile computer-readable storage medium is provided, on which computer program instructions are stored, wherein the computer program instructions implement the above-mentioned image processing method when executed by a processor.

[0023] According to another aspect of the present disclosure, a computer program product is provided, including a computer-readable code, or a non-volatile computer-readable storage medium carrying the computer-readable code. When the computer-readable code runs in a processor of an electronic device, the processor in the electronic device executes the above-mentioned image processing method.

[0024] Further features and aspects of the present disclosure will become apparent from the following detailed description of exemplary embodiments with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate exemplary embodiments, features, and aspects of the disclosure and, together with the description, serve to explain the principles of the disclosure.

[0026] Figures 1 to 3 A schematic diagram showing an image processing method provided according to an embodiment of the present disclosure.

[0027] Figures 4 to 5 A block diagram of an image processing device provided according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0028] Various exemplary embodiments, features and aspects of the present disclosure will be described in detail below with reference to the accompanying drawings. The same reference numerals in the accompanying drawings represent elements with the same or similar functions. Although various aspects of the embodiments are shown in the accompanying drawings, the drawings are not necessarily drawn to scale unless otherwise specified.

[0029] The word “exemplary” is used exclusively herein to mean “serving as an example, example, or illustration.” Any embodiment described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments.

[0030] In addition, in order to better illustrate the present disclosure, numerous specific details are given in the following specific embodiments. It should be understood by those skilled in the art that the present disclosure can also be implemented without certain specific details. In some examples, methods, means, components and circuits well known to those skilled in the art are not described in detail in order to highlight the subject matter of the present disclosure.

[0031] In order to facilitate those skilled in the art to understand the technical solution provided by the embodiments of the present disclosure, the technical environment in which the technical solution is implemented is described below.

[0032] When the speed of a high-speed moving object exceeds the limit that the image sensor can clearly capture, the image sensor will receive light information from different positions within a single exposure cycle, and this information is superimposed on the image to form an afterimage. This afterimage not only reduces the clarity of the image, but more seriously, it will lead to a significant decrease in measurement accuracy, because the afterimage makes the key features such as edges and contours in the image blurred, making it difficult to accurately extract and identify. When using a quantum dot spectral sensor to collect spectral images, if the object corresponding to the spectral image is a high-speed moving object, then because the movement distance of the high-speed moving object during the acquisition exposure time is greater than the spatial resolution distance of the image sensor, afterimages will be generated on the acquired spectral image, affecting the measurement accuracy, thereby affecting the accurate acquisition of image information.

[0033] In order to solve the above technical problems, an embodiment of the present disclosure provides an image processing method, including using the same quantum dot spectrum sensor to perform spectrum acquisition on a target area to obtain multiple first spectrum images, wherein the quantum dot spectrum sensor is provided with multiple quantum dot areas, each quantum dot area has an optical characteristic corresponding to a different spectrum channel, and each first spectrum image includes spectrum information under all spectrum channels; according to a preset image extraction rule, image extraction is performed on the multiple first spectrum images respectively to obtain an image set corresponding to each spectrum channel, each image set includes multiple extracted images of the corresponding spectrum channel, all extracted images in the same image set come from different first spectrum images, and each extracted image in the same image set has the same position in the corresponding first spectrum image; all extracted images in each image set are averaged to obtain an average image corresponding to each spectrum channel, and a second spectrum image is determined based on the average image corresponding to all spectrum channels, and the second spectrum image includes spectrum information under all spectrum channels after eliminating measurement errors, so that image afterimages in the spectrum image can be effectively eliminated while maintaining high measurement accuracy, so as to achieve accurate acquisition of spectrum images of high-speed moving objects using quantum dot spectrum sensors.

[0034] Figures 1 to 3 A schematic diagram of an image processing method provided according to an embodiment of the present disclosure is shown. Figures 1 to 3 The image processing method provided by the embodiment of the present disclosure is schematically illustrated.

[0035] like Figure 1 As shown, the image processing method may include the following steps S101 to S103.

[0036] Step S101: Spectrum collection is performed on a target area using the same quantum dot spectrum sensor to obtain a plurality of first spectrum images.

[0037] The quantum dot spectral sensor is provided with multiple quantum dot regions. Each quantum dot region has optical properties corresponding to different spectral channels. In other words, each quantum dot region has different optical properties, and each quantum dot region corresponds to a different spectral channel. Among them, quantum dots (QDs) are a nanoscale semiconductor crystal material with a radius close to or less than the exciton Bohr radius. Quantum dots can be used as light absorbing materials in image sensors. Depending on the quantum dot materials, the quantum dot regions can have different spectral response characteristics. A strip-shaped quantum dot array sheet with multiple quantum dot regions is placed in front of the detector, and the absorption characteristics of the spectrum of the quantum dot material are used to modulate the incident spectrum of the target to be measured, so that the corresponding spectral information can be obtained. In this way, by using the quantum dot spectral sensor to collect spectral images (or spectral collection), it has the advantages of high spectral resolution, high energy utilization, small size, wide spectral range and low cost.

[0038] The target area can be a preset area where the target to be measured can be obtained, so that the quantum dot spectrum sensor can be fixed at a position where the spectrum image of the target area can be obtained, wherein the target to be measured is a high-speed moving object, such as mud in water. Therefore, the quantum dot spectrum sensor can be used to collect the spectrum of the target area to obtain multiple first spectrum images with different collection times. Each first spectrum image includes spectrum information under all spectrum channels, that is, each first spectrum image includes sub-images under all spectrum channels, and each sub-image reflects the spectrum information of the corresponding spectrum channel. The number of spectrum channels is the same as the number of quantum dot areas. The actual position of the quantum dot area in the first spectrum image can be known in advance.

[0039] Step S102: performing image extraction on the plurality of first spectral images respectively according to a preset image extraction rule to obtain an image set corresponding to each spectral channel.

[0040] Multiple image sets can be obtained through step S102. Different image sets correspond to different spectral channels. Each image set includes multiple extracted images corresponding to the spectral channels. All extracted images in the same image set come from different first spectral images. The extracted images in the same image set have the same position in the corresponding first spectral image. For example, three first spectral images A1, A2, and A3 are obtained through spectral acquisition. The first spectral images A1, A2, and A3 all include sub-images of the three spectral channels B1, B2, and B3. By performing image extraction on the first spectral images A1, A2, and A3, three sets can be obtained, namely, image set C1 corresponding to spectral channel B1, image set C2 corresponding to spectral channel B2, and image set C3 corresponding to spectral channel B3. Image sets C1, C2, and C3 all include three extracted images. Image set C1 includes The extracted images D1, D2, and D3 are included. If the upper left corner area of ​​the first spectral image is selected as the extracted area, the extracted image D1 in the image set C1 is the upper left corner area of ​​the sub-image corresponding to the spectral channel B1 in the first spectral image A1, the extracted image D2 is the upper left corner area of ​​the sub-image corresponding to the spectral channel B1 in the first spectral image A2, and the extracted image D3 is the upper left corner area of ​​the sub-image corresponding to the spectral channel B1 in the first spectral image A3. The image sets C2 and C3 are similar to C1 and are not repeated here.

[0041] The extraction area related to the extracted image can be determined by the preset image extraction rules. Image extraction rules can be divided into clear position extraction rules and fixed position extraction rules. The clear position extraction rule is used to extract the clearer part of the image. The fixed position extraction rule is used to extract the fixed part of the image. Both the clear part and the fixed part should cover all spectral channels. This is because the spectral image obtained by spectral acquisition has information of multiple spectral channels. In order to eliminate the measurement error on each spectral channel, it is necessary to perform image extraction on each spectral channel. Therefore, no matter how the relevant extraction area is set or obtained, it should cover all spectral channels.

[0042] In the case where the extraction area is a continuous area, step S102 may include: using the image in the target full spectrum extraction area in each first spectrum image as the first image corresponding to each first spectrum image; dividing each first image based on the position information of each quantum dot area to obtain multiple first sub-images, each first sub-image of the same first image corresponds to a different spectral channel; and determining the image set corresponding to each spectral channel based on all first sub-images. Wherein, the target full spectrum extraction area is a continuous area covering all spectral channels.

[0043] A default full spectrum extraction area can be preset in advance as the above continuous area. In each first spectrum image, image extraction can be performed based on the preset default full spectrum extraction area to obtain the corresponding first image. Then, in each first image, segmentation can be performed based on the position information of each quantum dot area to obtain a corresponding segmentation result, and the segmentation result includes multiple first sub-images. Finally, the image set corresponding to each spectral channel can be determined based on all segmentation results, that is, all first sub-images.

[0044] Now Figure 2 The first spectral image A1 including spectral information of three spectral channels B1, B2, and B3 is taken as an example to illustrate the image extraction process based on the default full spectrum extraction area: Figure 2 The area corresponding to the red dotted box is the preset default full spectrum extraction area Q. The default full spectrum extraction area Q covers the three spectral channels B1, B2, and B3. Figure 2As shown, an image corresponding to the default full spectrum extraction area Q can be extracted from the first spectrum image A1, that is, the first image corresponding to the first spectrum image A1, and the first image includes information of three spectrum channels, namely, spectrum channels B1, B2, and B3. Since all the extracted images in an image set correspond to the same spectrum channel, the first image needs to be segmented for each spectrum channel to obtain the corresponding segmentation result, that is, the first image is segmented into three first sub-images according to the position information of each quantum dot region, namely, the first sub-image IMG1 corresponding to the spectrum channel B1, the first sub-image IMG2 corresponding to the spectrum channel B2, and the first sub-image IMG3 corresponding to the spectrum channel B3, so that the first sub-image IMG1 is used as one of the extracted images in the image set corresponding to the spectrum channel B1, the first sub-image IMG2 is used as one of the extracted images in the image set corresponding to the spectrum channel B2, and the first sub-image IMG3 is used as one of the extracted images in the image set corresponding to the spectrum channel B3. The image extraction process of other first spectrum images is similar to this, so that the image extraction is realized through the above process, and the image set corresponding to each spectrum channel can be finally obtained.

[0045] In addition to the preset default full spectrum extraction area, a continuous area can also be determined based on the clarity of the first spectrum image actually obtained. Based on this, the image processing method may also include: determining the clarity of each first spectrum image, and the calculation method of the clarity can be flexibly set according to the actual situation; when the number of first spectrum images with clarity greater than or equal to the clarity threshold is zero, the preset default full spectrum extraction area is determined as the target full spectrum extraction area. In other words, in this case, the clarity of all first spectrum images obtained by spectrum acquisition does not meet expectations, so the preset default full spectrum extraction area can be directly used as the target full spectrum extraction area. The specific extraction process is detailed in the previous text and will not be repeated here. The specific value of the clarity threshold can be flexibly set according to the calculation method of the clarity and the actual situation; when the number of first spectrum images with clarity greater than or equal to the clarity threshold is not zero, the selected full spectrum extraction area determined based on the first spectrum image with the highest clarity is determined as the target full spectrum extraction area. In other words, in this case, the clarity of at least one first spectrum image among all first spectrum images obtained by spectrum acquisition meets expectations, so the target full spectrum extraction area can be determined based on the first spectrum image with the highest clarity.

[0046] To achieve the above process of determining the target full spectrum extraction area, the image processing method may further include a process of determining the number of first spectrum images whose clarity is greater than or equal to a clarity threshold and a process of determining a selected full spectrum extraction area.

[0047] For example, the process of determining the number of first spectral images whose clarity is greater than or equal to a clarity threshold may include the following steps: converting each first spectral image into a corresponding grayscale image, the size of the grayscale image being the same as the size of the corresponding first spectral image; performing Laplace transform on each grayscale image to obtain a corresponding transformed image, the size of the transformed image being the same as the size of the corresponding grayscale image, wherein Laplace transform refers to adjusting the pixel value at the target position using the pixel value at the target position and multiple pixel values ​​in its vicinity; calculating the variance value corresponding to each transformed image, the variance value can represent the clarity of the image; comparing each variance value with a preset clarity threshold to obtain a first number of transformed images whose variance value is greater than the clarity threshold and a second number of transformed images whose variance value is equal to the clarity threshold, and taking the sum of the first number and the second number as the number of first spectral images whose clarity is greater than or equal to the clarity threshold.

[0048] Still taking the variance value to represent the clarity of the image as an example, the process of determining the selected full spectrum extraction area is explained: when the number of first spectrum images whose clarity is greater than or equal to the clarity threshold is not zero, determine the transformed image with the largest variance value; select a selected full spectrum extraction area covering all spectral channels from the transformed image, wherein the selection method can be flexibly set according to actual needs. When the selected full spectrum extraction area is determined, image extraction can be performed in each first spectrum image based on the selected full spectrum extraction area to obtain a corresponding first image. Then, in each first image, segmentation can be performed based on the position information of each quantum dot area to obtain a corresponding segmentation result, and the segmentation result includes multiple first sub-images. Finally, the image set corresponding to each spectral channel can be determined based on all segmentation results, i.e., all first sub-images.

[0049] In the case where the extraction region is a plurality of sub-regions, step S102 may include: using the images in each target spectral channel extraction region in each first spectral image as a plurality of second sub-images corresponding to each first spectral image; and determining the image set corresponding to each spectral channel based on all the second sub-images. Wherein, one target spectral channel extraction region corresponds to one spectral channel, and the positions of different target spectral channel extraction regions in the corresponding spectral channel may be the same or different, and all target spectral channel extraction regions cover all spectral channels.

[0050] Multiple default spectral channel extraction regions can be preset in advance as the multiple sub-regions. In any first spectral image, the images corresponding to each target spectral channel extraction region can be extracted to obtain multiple second sub-images corresponding to the first spectral image, one second sub-image corresponds to one spectral channel, and all second sub-images cover all spectral channels. All second sub-images can be obtained through such an extraction process, and according to the division method of the spectral channel, an image set corresponding to the spectral channel is formed according to all second sub-images corresponding to the same spectral channel, so that an image set corresponding to each spectral channel can be obtained.

[0051] Now Figure 3 The first spectral image A1 including spectral information of three spectral channels B1, B2, and B3 is taken as an example to illustrate the image extraction process based on the image set corresponding to each default spectral channel extraction area: Figure 3 The areas corresponding to the three red dotted boxes are the preset default spectral channel extraction areas P1, P2, and P3. The default spectral channel extraction area P1 corresponds to spectral channel B1, the default spectral channel extraction area P2 corresponds to spectral channel B2, and the default spectral channel extraction area P3 corresponds to spectral channel B3. Figure 3 As shown, the image in the default spectral channel extraction area P1 in the first spectral image A1 can be extracted and used as the second sub-image E1 corresponding to the spectral channel B1. Similarly, the second sub-image E2 corresponding to the spectral channel B2 and the second sub-image E3 corresponding to the spectral channel B3 can be obtained, so that the second sub-image E1 is used as one of the extracted images in the image set corresponding to the spectral channel B1, the second sub-image E2 is used as one of the extracted images in the image set corresponding to the spectral channel B2, and the second sub-image E3 is used as one of the extracted images in the image set corresponding to the spectral channel B3. The image extraction process of other first spectral images is similar to this. In this way, image extraction is achieved through the above process, and finally the image set corresponding to each spectral channel can be obtained.

[0052] In addition to the preset default spectral channel extraction areas, the above-mentioned multiple sub-areas can also be determined based on the clarity of the first spectral image actually obtained. Based on this, the image processing method can also include: determining the clarity of each first spectral image, the calculation method of the clarity and the description of the clarity threshold are described in the previous text, which will not be repeated here; when the number of first spectral images with clarity greater than or equal to the clarity threshold is zero, the preset default spectral channel extraction areas are determined as the target spectral channel extraction areas. In other words, in this case, the clarity of all first spectral images obtained by spectral acquisition does not meet expectations, so the preset default spectral channel extraction areas can be directly used as the target spectral channel extraction areas. The specific extraction process is detailed in the previous text, which will not be repeated here; when the number of first spectral images with clarity greater than or equal to the clarity threshold is not zero, the selected spectral channel extraction areas determined based on the first spectral image with the highest clarity are used as the target spectral channel extraction areas. In other words, in this case, the clarity of at least one first spectral image among all first spectral images obtained by spectral acquisition meets expectations, so the target spectral channel extraction areas can be determined based on the first spectral image with the highest clarity. The process of determining the number of first spectrum images whose clarity is greater than or equal to the clarity threshold is as described above and will not be repeated here.

[0053] To achieve the above process of determining each target spectral channel extraction area, the image processing method may also include a process of determining each selected spectral channel extraction area: when the number of first spectral images with clarity greater than or equal to the clarity threshold is not zero, determine the transformed image with the largest variance value; based on the position information of each quantum dot area, divide the transformed image into blocks to obtain multiple sub-blocks corresponding to each spectral channel, and the sub-blocks corresponding to one spectral channel are of the same size; calculate the clarity evaluation value corresponding to each sub-block under each spectral channel to obtain the sub-block with the highest clarity evaluation value under each spectral channel, wherein the clarity evaluation value may be an energy gradient function (EOG) value, or other functions may be selected for calculation, which may be flexibly set according to actual needs; and use the sub-block with the highest clarity evaluation value under each spectral channel as each selected spectral channel extraction area. When each selected spectral channel extraction area is determined, in any first spectral image, the image corresponding to each target spectral channel extraction area may be extracted to obtain multiple second sub-images corresponding to the first spectral image, and according to the division method of the spectral channel, the image set corresponding to the spectral channel is formed according to all the second sub-images corresponding to the same spectral channel, so that the image set corresponding to each spectral channel can be obtained.

[0054] The image processing method may further include: determining each first spectral image and / or each extracted image as an image to be corrected, and performing image correction on each image to be corrected. In other words, before executing step S102, image correction may be performed on each first spectral image, and then image extraction may be performed on the corrected first spectral image. Alternatively, after executing step S102, image correction may be performed on each extracted image in each image set, and then all corrected extracted images may be averaged. Alternatively, image correction may be performed both before and after image extraction.

[0055] In addition to the quantum dot area, the quantum dot spectral sensor is also provided with a reference area. The reference area is an area where quantum dots are not arranged or the concentration of the arranged quantum dots is less than the preset concentration threshold. In the process of image correction, the second signal corresponding to the reference area can be used to correct the first signal corresponding to the quantum dot area. The above-mentioned preset concentration threshold can be set according to actual needs. Generally, the concentration of quantum dots set in the quantum dot area will be greater than the preset concentration threshold.

[0056] When the spatial distribution of the intensity of the light of the light source (hereinafter referred to as the spatial distribution of light intensity) is uniform, a global correction can be used to correct the spatial distribution of the intensity. When the spatial distribution of light intensity corresponding to the spectral acquisition meets the preset uniform distribution condition, the global Gaussian distribution is determined based on the reference area, and the global Gaussian distribution is used to correct the first signal corresponding to the quantum dot area to obtain a corrected image. The global correction uses all the second areas to fit a global Gaussian spot shape distribution, so as to obtain the spatial distribution of the intensity of the incident light at all spectral channels, and then deduct the spatial distribution of light intensity from the first signal to obtain the corrected image. In this way, the use of global correction can eliminate or reduce the image distortion and noise caused by the uneven spatial distribution of the light source intensity, making the corrected image clearer and more accurate.

[0057] When the spatial distribution of light intensity is uneven, local correction can be used to correct the spatial distribution of intensity. When the spatial distribution of light intensity meets the preset non-uniform conditions, interpolation processing is performed based on the second signal corresponding to the reference area and the first signal corresponding to the quantum dot area to obtain a corrected image. Local correction uses the spatial distribution of light intensity in the reference area near a spectral channel to interpolate the spatial distribution of intensity at the spectral channel (quantum dot area), thereby deducting the spatial distribution of light intensity from the first signal to obtain a corrected image. In this way, there is no need to make large-scale adjustments to the entire system, but only to calibrate specific spectral channels, which greatly simplifies the correction process and reduces correction costs and time.

[0058] Alternatively, when the spatial distribution of light intensity is uneven, especially extremely uneven, a high-order correction method can be used for correction. That is, when the spatial distribution of light intensity meets the non-uniform condition, a high-order correction is performed based on the reference area and the quantum dot area to obtain a corrected image. High-order correction can use more complex mathematical models or algorithms, such as polynomial fitting, bilinear interpolation, surface fitting, etc., to achieve the deduction of information indicating the extremely uneven spatial distribution of light intensity from the first signal to obtain a corrected image. In this way, the use of high-order correction can handle more complex light intensity distribution situations and more accurately predict the spatial distribution of light intensity, making the corrected image clearer and more accurate.

[0059] Step S103: average all extracted images in each image set to obtain an average image corresponding to each spectral channel, and determine a second spectral image based on the average images corresponding to all spectral channels.

[0060] The second spectral image includes spectral information under all spectral channels after eliminating the measurement error. The second spectral image after eliminating the measurement error can accurately reflect the relevant information of the high-speed moving object, which is helpful for the subsequent calculation of specific data of the high-speed moving object. For example, if the second spectral image reflects the information of sediment in the water, the sediment content can be accurately calculated based on the second spectral image.

[0061] The averaging process in step S103 may be to directly sum and then average. Based on this, the averaging process of all the extracted images in each image set in step S103 to obtain the average image corresponding to each spectral channel may include: for each image set, the average value of each pixel position is calculated based on the pixel values ​​of all the extracted images in the image set at the same pixel position, and the average image of the corresponding spectral channel is determined according to the average value of all pixel positions. Take the image set C1 corresponding to the spectral channel B1 as an example, which includes the extracted images D1, D2, and D3, and D1, D2, and D3 all include 4 pixel positions, to illustrate the averaging process: Since D1 to D3 all include 4 pixel positions, the corresponding average image also includes 4 pixel positions. In this example, L1 to L4 are used to represent these 4 pixel positions. According to the pixel value at the pixel position L1 in D1 to D3, the average value X1 corresponding to L1 is calculated, and the average value X1 is used as the pixel value at the pixel position L1 in the average image. The remaining pixel positions are similarly calculated, thereby calculating the average image corresponding to the spectral channel B1. The average processing of the remaining image sets is similar to this and will not be described in detail in this article.

[0062] Alternatively, the averaging process in step S103 may also be a single pixel calibration weighted average, that is, an average process combined with an adjustment process based on a preset pixel reference value. Based on this, the averaging process of all extracted images in each image set in step S103 to obtain an average image corresponding to each spectral channel may include: for each image set, calculating the average value of each pixel position based on the pixel values ​​of all extracted images in the image set at the same pixel position, and determining the average image of the corresponding spectral channel based on the average value adjusted by the corresponding pixel reference value. The averaging process is still described by taking the image set C1 corresponding to the spectral channel B1 including the extracted images D1, D2, and D3, and D1, D2, and D3 each including 4 pixel positions as an example: the description of each pixel position of the average image and the average value calculation process refer to the previous average process example, which will not be repeated here. After determining the average values ​​X1 to X4 corresponding to the pixel positions L1 to L4, the preset corresponding pixel reference values ​​R1 to R4 can be used to adjust X1 to X4, that is, the sum of X1 and R1 can be used as the pixel value at the pixel position L1 in the average image, and the rest of the pixel positions are similar to this, thereby calculating the average image corresponding to the spectral channel B1. The average processing process of the rest of the image sets is similar to this, and will not be repeated in this article.

[0063] Determining the second spectral image based on the average image corresponding to all spectral channels in step S103 may include: splicing the average images of the corresponding spectral channels in a predetermined order of the spectral channels to obtain the second spectral image. The order of the spectral channels may be determined according to the position information of each quantum dot region. For example, the quantum dot array sheet in the image sensor has 7 quantum dot regions. Accordingly, the first spectral image includes information of 7 spectral channels B1 to B7 distributed from left to right. After obtaining the average image corresponding to the 7 spectral channels, the average images corresponding to the spectral channels B1 to B7 may be spliced ​​in order from left to right to obtain the second spectral image.

[0064] Before executing step S103, the extracted images in each image set may be cleaned to remove abnormal images and improve the accuracy of the subsequent second spectral image. Based on this, the image processing method may also include: performing abnormality detection on the pixel values ​​at all positions in each extracted image to obtain the detection results of each extracted image, the detection results represent the number of abnormal pixel values ​​in the extracted image, and the abnormality detection is performed based on a preset pixel interval; based on the detection results of each extracted image, deleting or correcting the abnormal extracted image. For example, if the pixel value of a certain position in the extracted image is in a preset pixel interval, the pixel value of the position is determined to be a normal pixel value. If the pixel value of the position is not in the pixel interval, the pixel value of the position is determined to be an abnormal pixel value. The abnormal detection of the remaining positions of the extracted image is similar to this, thereby obtaining the detection result of the extracted image, that is, the number of abnormal pixel values. If the number is greater than the preset threshold, the extracted image is deleted from the image set in which it is located. If the number is less than or equal to the preset threshold, the abnormal pixel points in the extracted image are corrected. The processing process of the remaining extracted images is similar to this, thereby obtaining each image set after cleaning, and the average processing can be performed based on the cleaned image set to obtain a second spectral image. In this way, by detecting and processing abnormal pixel values, noise, errors or pixels that do not meet expectations in the image can be effectively removed, thereby significantly improving the overall quality of the image. In addition, by detecting and processing abnormal pixel values, the image can be deleted or corrected, which can reduce the computational burden and time cost of subsequent processing steps. Especially in large-scale image processing scenarios, this optimization is particularly critical and can significantly improve the overall processing efficiency.

[0065] It should be noted that, in addition to the above-mentioned cleaning process, the image can also be cleaned by 3σ detection, box plot and other methods, which are not limited in the embodiments of the present disclosure. In addition, the image processing method may also include denoising the image. Denoising can be performed by median filtering, Gaussian filtering and other methods. The object of denoising can be at least one of the first spectral image, the extracted image and the second spectral image. In fact, in the entire image processing process, the image can be denoised, and the embodiments of the present disclosure do not limit the denoising object, the number of denoising times and the denoising method.

[0066] After determining the second spectral image after eliminating the error, a mapping relationship can be established based on the real spectral signal and the second spectral image, so as to eliminate the error of the spectral image measured subsequently. Based on this, the image processing method can also include: constructing a target mapping relationship representing the error between the real spectral information and the measured spectral information based on the real spectral image and the second spectral image corresponding to the target area. In this way, the error introduced by factors such as the measuring equipment, environmental conditions or data processing can be significantly reduced based on the target mapping relationship, thereby improving the accuracy of the spectral measurement; and the establishment of the target mapping relationship is based on the real spectral image and the second spectral image after the error is eliminated, which ensures the accuracy and reliability of the mapping relationship. Based on this mapping relationship, the subsequent spectral images can be processed to generate a spectral image that is closer to the actual situation, providing a more reliable data basis for the subsequent image analysis; at the same time, once the error mapping relationship is established, the error elimination processing of the spectral image measured subsequently can be automatically performed without manual intervention, which not only improves the processing efficiency, but also reduces the possibility of human error, making the processing of spectral data standardized and repeatable.

[0067] This image processing method is suitable for measuring the balance properties of a high-speed quasi-dynamic balance system. For a high-speed quasi-dynamic balance system, although the instantaneous state recorded by a single spectral image has a high degree of random uncertainty, the result of multiple measurements and averages by this image processing method can obtain a spectral image with an expected value that tends to the probability distribution, that is, the balance property of the quasi-dynamic balance system, eliminating the measurement error caused by motion afterimages while maintaining high measurement accuracy.

[0068] The embodiment of the present disclosure also provides an image processing device. Figure 4 FIG. 1 is a block diagram of an image processing device according to an embodiment of the present disclosure. Figure 4 As shown, the image processing device 100 includes the following spectrum acquisition module 101, image extraction module 102, and average processing module 103.

[0069] The spectral acquisition module 101 is used to use the same quantum dot spectral sensor to perform spectral acquisition on the target area to obtain multiple first spectral images, wherein the quantum dot spectral sensor is provided with multiple quantum dot areas, each of the quantum dot areas has optical characteristics corresponding to different spectral channels, and each of the first spectral images includes spectral information under all spectral channels.

[0070] The image extraction module 102 is used to perform image extraction on the multiple first spectral images respectively according to preset image extraction rules to obtain image sets corresponding to each spectral channel, each of the image sets includes multiple extracted images of the corresponding spectral channel, all extracted images in the same image set come from different first spectral images, and the positions of the extracted images in the same image set in the corresponding first spectral image are the same.

[0071] The averaging processing module 103 is used to perform averaging processing on all the extracted images in each of the image sets to obtain an average image corresponding to each of the spectral channels, and determine a second spectral image based on the average images corresponding to all spectral channels, wherein the second spectral image includes spectral information under all spectral channels after eliminating measurement errors.

[0072] In a possible implementation, according to a preset image extraction rule, image extraction is performed on the multiple first spectral images respectively to obtain an image set corresponding to each spectral channel, including: using the image in the target full-spectrum extraction area in each of the first spectral images as the first image corresponding to each first spectral image, and the target full-spectrum extraction area is a continuous area covering all spectral channels; obtaining multiple first sub-images by segmenting each of the first images based on the position information of each of the quantum dot areas; determining the image set corresponding to each of the spectral channels based on all the first sub-images, and the spectral channels corresponding to each first sub-image of the same first image are different; or using the image in each target spectral channel extraction area in each of the first spectral images as the multiple second sub-images corresponding to each first spectral image; and determining the image set corresponding to each of the spectral channels based on all the second sub-images.

[0073] In a possible implementation, the device also includes a determination module, which is used to: determine the clarity of each of the first spectral images; when the number of first spectral images with clarity greater than or equal to a clarity threshold is zero, determine a preset default full-spectrum extraction area as the target full-spectrum extraction area; when the number of first spectral images with clarity greater than or equal to a clarity threshold is not zero, determine a selected full-spectrum extraction area determined based on the first spectral image with the highest clarity as the target full-spectrum extraction area.

[0074] In a possible implementation manner, the device further includes a correction module, which is used to: determine each of the first spectral images and / or each of the extracted images as an image to be corrected, and perform image correction on each of the images to be corrected.

[0075] In a possible implementation, the quantum dot spectral sensor is further provided with a reference area, in which no quantum dots are deployed or the concentration of the deployed quantum dots is less than a preset concentration threshold; the image correction includes at least one of the following: when the spatial distribution of light intensity corresponding to the spectral acquisition satisfies a preset uniform distribution condition, a global Gaussian distribution is determined based on the reference area, and the global Gaussian distribution is used to correct the first signal corresponding to the quantum dot area to obtain a corrected image; when the spatial distribution of light intensity satisfies a preset non-uniform distribution condition, interpolation processing is performed based on the first signal corresponding to the quantum dot area and the second signal corresponding to the reference area, or high-order correction is performed based on the quantum dot area and the reference area to obtain a corrected image.

[0076] In a possible implementation, the device also includes an anomaly detection module, which is used to: perform anomaly detection on pixel values ​​at all positions in each of the extracted images to obtain a detection result of each of the extracted images, wherein the detection result represents the number of abnormal pixel values ​​in the extracted image, and the anomaly detection is performed based on a preset pixel interval; based on the detection result of each of the extracted images, delete or correct the abnormal extracted image.

[0077] In a possible implementation, all the extracted images in each of the image sets are averaged to obtain an average image corresponding to each of the spectral channels, including: for each of the image sets, the average value of each pixel position is calculated based on the pixel values ​​of all the extracted images in the image set at the same pixel position, and the average image of the corresponding spectral channel is determined according to the average value of all pixel positions or the average value adjusted based on the corresponding pixel reference value; and / or, a second spectral image is determined based on the average image corresponding to all spectral channels, including: according to a predetermined order of the spectral channels, the average images of the corresponding spectral channels are spliced ​​to obtain the second spectral image.

[0078] In a possible implementation manner, the device further includes a construction module, which is used to construct a target mapping relationship representing an error between real spectral information and measured spectral information based on the real spectral image corresponding to the target area and the second spectral image.

[0079] In some embodiments, the functions or modules included in the device provided by the embodiments of the present disclosure can be used to execute the method described in the above method embodiments. The specific implementation can refer to the description of the above method embodiments, and for the sake of brevity, it will not be repeated here.

[0080] The disclosed embodiment also provides a method for measuring sediment content, including: performing calculations based on a second spectral image determined using the above-mentioned image processing method to obtain the sediment content in a target area corresponding to the second spectral image.

[0081] The present disclosure also provides a computer-readable storage medium on which computer program instructions are stored, and when the computer program instructions are executed by a processor, at least one of the above-mentioned image processing method and sediment content measurement method is implemented. The computer-readable storage medium can be a volatile or non-volatile computer-readable storage medium.

[0082] The embodiment of the present disclosure also proposes an electronic device, comprising: a processor; a memory for storing instructions executable by the processor; wherein the processor is configured to implement at least one of the above-mentioned image processing method and sediment content measurement method when executing the instructions stored in the memory.

[0083] The embodiments of the present disclosure also provide a computer program product, including a computer-readable code, or a non-volatile computer-readable storage medium carrying the computer-readable code. When the computer-readable code runs in a processor of an electronic device, the processor in the electronic device executes at least one of the above-mentioned image processing method and sediment content measurement method.

[0084] Figure 5 5 shows a block diagram of an image processing apparatus according to an embodiment of the present disclosure. For example, the apparatus 500 may be provided as a server or a terminal device. Figure 5 , the apparatus 500 includes a processing component 522, which further includes one or more processors, and a memory resource represented by a memory 532 for storing instructions, such as an application, that can be executed by the processing component 1922. The application stored in the memory 1932 may include one or more modules, each corresponding to a set of instructions. In addition, the processing component 1922 is configured to execute instructions to perform the above method.

[0085] The device 1900 may also include a power supply component 1926 configured to perform power management of the device 1900, a wired or wireless network interface 1950 configured to connect the device 1900 to a network, and an input / output interface 1958 (I / O interface). The device 1900 may operate based on an operating system stored in the memory 1932, such as Windows Server 2000. TM , MacOS X TM , Unix TM ,Linux TM , FreeBSD TM or similar.

[0086] In an exemplary embodiment, a non-volatile computer-readable storage medium is also provided, such as a memory 1932 including computer program instructions, which can be executed by the processing component 1922 of the device 1900 to perform the above method.

[0087] The present disclosure may be a system, a method and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present disclosure.

[0088] A computer-readable storage medium may be a tangible device that can hold and store instructions used by an instruction execution device. A computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples of computer-readable storage media (a non-exhaustive list) include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination of the foregoing. As used herein, a computer-readable storage medium is not to be interpreted as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through a wire.

[0089] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing / processing device.

[0090] The computer program instructions for performing the operation of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages, such as Smalltalk, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. Computer-readable program instructions may be executed completely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or completely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., using an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), may be customized by utilizing the state information of the computer-readable program instructions, and the electronic circuit may execute the computer-readable program instructions, thereby realizing various aspects of the present disclosure.

[0091] Various aspects of the present disclosure are described herein with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer-readable program instructions.

[0092] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processor of the computer or other programmable data processing device, a device that implements the functions / actions specified in one or more boxes in the flowchart and / or block diagram is generated. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause the computer, programmable data processing device, and / or other equipment to work in a specific manner, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0093] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operating steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0094] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to multiple embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and a part of the module, program segment or instruction includes one or more executable instructions for realizing the specified logical function. In some alternative implementations, the function marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or action, or can be implemented with a combination of special hardware and computer instructions.

[0095] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or technical improvements in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. An image processing method, characterized in that: include: Using the same quantum dot spectrum sensor to collect spectrum for the target area, a plurality of first spectrum images are obtained, wherein the quantum dot spectrum sensor is provided with a plurality of quantum dot regions, each of the quantum dot regions has an optical characteristic corresponding to a different spectrum channel, and each of the first spectrum images includes spectrum information under all spectrum channels; According to a preset image extraction rule, image extraction is performed on the multiple first spectral images respectively to obtain image sets corresponding to the spectral channels, each of the image sets includes multiple extracted images of the corresponding spectral channels, all the extracted images in the same image set come from different first spectral images, and the positions of the extracted images in the same image set in the corresponding first spectral image are the same; All extracted images in each of the image sets are averaged to obtain an average image corresponding to each of the spectral channels, and a second spectral image is determined based on the average images corresponding to all spectral channels, wherein the second spectral image includes spectral information under all spectral channels after eliminating measurement errors.

2. The method according to claim 1, characterized in that According to a preset image extraction rule, image extraction is performed on the plurality of first spectral images respectively to obtain an image set corresponding to each spectral channel, including: Using the image in the target full spectrum extraction area in each of the first spectrum images as the first image corresponding to each first spectrum image, the target full spectrum extraction area is a continuous area covering all spectral channels; dividing each of the first images based on the position information of each of the quantum dot regions to obtain a plurality of first sub-images; determining an image set corresponding to each of the spectral channels based on all the first sub-images, wherein each first sub-image of the same first image corresponds to a different spectral channel; or The images in the target spectral channel extraction areas in the first spectral images are used as a plurality of second sub-images corresponding to the first spectral images; and the image sets corresponding to the spectral channels are determined based on all the second sub-images.

3. The method according to claim 2, characterized in that The method further comprises: Determining the clarity of each of the first spectral images; When the number of first spectrum images whose clarity is greater than or equal to the clarity threshold is zero, determining a preset default full spectrum extraction area as the target full spectrum extraction area; When the number of first spectrum images with clarity greater than or equal to the clarity threshold is not zero, the selected full spectrum extraction region determined based on the first spectrum image with the highest clarity is determined as the target full spectrum extraction region.

4. The method according to any one of claims 1 to 3, characterized in that The method further comprises: Each of the first spectral images and / or each of the extracted images is determined as an image to be corrected, and image correction is performed on each of the images to be corrected.

5. The method according to claim 4, characterized in that The quantum dot spectral sensor is further provided with a reference area, in which no quantum dots are arranged or the concentration of the arranged quantum dots is less than a preset concentration threshold; the image correction includes at least one of the following: In the case where the spatial distribution of light intensity corresponding to the spectrum acquisition satisfies a preset uniform distribution condition, a global Gaussian distribution is determined based on the reference area, and the first signal corresponding to the quantum dot area is corrected using the global Gaussian distribution to obtain a corrected image; When the spatial distribution of light intensity satisfies a preset non-uniform distribution condition, interpolation processing is performed based on the first signal corresponding to the quantum dot area and the second signal corresponding to the reference area, or high-order correction is performed based on the quantum dot area and the reference area to obtain a corrected image.

6. The method according to any one of claims 1 to 3, characterized in that: The method further comprises: Performing anomaly detection on pixel values ​​at all positions in each of the extracted images to obtain a detection result of each of the extracted images, wherein the detection result indicates the number of abnormal pixel values ​​in the extracted image, and the anomaly detection is performed based on a preset pixel interval; Based on the detection result of each of the extracted images, abnormal extracted images are deleted or corrected.

7. The method according to any one of claims 1 to 3, characterized in that Averaging all the extracted images in each of the image sets to obtain an average image corresponding to each of the spectral channels includes: For each of the image sets, an average value of each pixel position is calculated based on the pixel values ​​of all extracted images in the image set at the same pixel position, and an average image of the corresponding spectral channel is determined based on the average value of all pixel positions or the average value adjusted based on the corresponding pixel reference value; and / or, Determining a second spectral image based on average images corresponding to all spectral channels includes: According to the predetermined order of the spectral channels, the average images of the corresponding spectral channels are spliced ​​to obtain the second spectral image.

8. The method according to any one of claims 1 to 3, characterized in that The method further comprises: Based on the real spectral image corresponding to the target area and the second spectral image, a target mapping relationship representing the error between the real spectral information and the measured spectral information is constructed.

9. An image processing device, characterized in that: include: A spectrum acquisition module, used for acquiring spectrum of a target area using the same quantum dot spectrum sensor to obtain a plurality of first spectrum images, wherein the quantum dot spectrum sensor is provided with a plurality of quantum dot regions, each of the quantum dot regions has an optical characteristic corresponding to a different spectrum channel, and each of the first spectrum images includes spectrum information under all spectrum channels; An image extraction module, configured to perform image extraction on the plurality of first spectral images respectively according to a preset image extraction rule, to obtain an image set corresponding to each spectral channel, wherein each image set includes a plurality of extracted images of the corresponding spectral channel, all extracted images in the same image set come from different first spectral images, and each extracted image in the same image set has the same position in the corresponding first spectral image; The averaging processing module is used to perform averaging processing on all the extracted images in each of the image sets to obtain an average image corresponding to each of the spectral channels, and determine a second spectral image based on the average images corresponding to all spectral channels, wherein the second spectral image includes spectral information under all spectral channels after eliminating measurement errors.

10. A method for calculating sediment content, characterized in that: include: The calculation is performed based on the second spectral image determined by the image processing method according to any one of claims 1 to 8 to obtain the sediment content in the target area corresponding to the second spectral image.

Images