Method, device, storage medium and electronic device for determining artificial rainbow

Abstract

The present application discloses a method, device, storage medium, and electronic device for determining an artificial rainbow. The method includes: obtaining an original image through an image acquisition device, wherein the original image carries rainbow information; determining the rainbow information based on the original image and a target network framework, wherein the target network framework integrates: a target detection algorithm, a bilateral segmentation network, wherein the rainbow information includes: rainbow frame information, rainbow mask information; performing a quality assessment on the rainbow information to obtain an assessment result, wherein the quality assessment includes at least one of the following: color channel index assessment, edge clarity assessment, and curve integrity assessment; forming rainbow imaging effect feedback based on the assessment result, and adjusting the rainbow imaging based on the rainbow imaging effect feedback. Through this application, the problem in the related art that the rainbow image cannot be deployed and evaluated at the edge is solved.

Description

Technical Field

[0001] The present application relates to the field of rainbow imaging technology, and in particular to a method, device, storage medium and electronic device for determining an artificial rainbow. Background Art

[0002] In the field of artificial rainbow imaging, control solutions are all based on PC platforms. This results in extended data transmission time, slow overall system response, poor data privacy, and low artificial rainbow image quality during artificial rainbow imaging evaluation. Therefore, how to deploy and evaluate rainbow images on edge platforms, which are different from PC platforms, is an urgent problem that needs to be solved.

[0003] Currently, no effective solution has been proposed to the problem that rainbow images cannot be deployed and evaluated at the edge in related technologies. Summary of the Invention

[0004] The main purpose of this application is to provide a method, device, storage medium and electronic device for determining an artificial rainbow, so as to solve the problem in the related art that rainbow images cannot be deployed and evaluated at the edge.

[0005] To achieve the above-mentioned objectives, according to the first aspect of the present application, a method for determining an artificial rainbow is provided. The method comprises: acquiring an original image through an image acquisition device, wherein the original image carries rainbow information; determining the rainbow information based on the original image and a target network framework, wherein the target network framework integrates: a target detection algorithm and a bilateral segmentation network, wherein the rainbow information includes: rainbow frame information and rainbow mask information; performing a quality assessment on the rainbow information to obtain an assessment result, wherein the quality assessment includes at least one of the following: color channel index assessment, edge clarity assessment, and curve integrity assessment; generating rainbow imaging effect feedback based on the assessment result, and adjusting the rainbow imaging based on the rainbow imaging effect feedback.

[0006] Furthermore, rainbow information is determined based on the original image and the target network framework, including: inputting the original image into the target detection algorithm to obtain rainbow frame information; and using the rainbow frame information as the rainbow information.

[0007] Furthermore, rainbow information is determined based on the original image and the target network framework, including: obtaining feature map information of a preset number of layers of rainbow frame information in the target detection algorithm; inputting the feature map information into the bilateral segmentation network to obtain a rainbow segmented image; when it is identified that the number of pixels in the target area in the rainbow segmented image is greater than the preset number of pixels, the area with the largest number of pixels in the target area is used as rainbow mask information.

[0008] Furthermore, the quality assessment is a color channel index assessment, which performs a quality assessment on the rainbow information to obtain an assessment result, including: processing the pixel area in the rainbow frame information except the rainbow mask information as black to obtain the processed rainbow information; performing spatial conversion on the processed rainbow information to output the color channel index of the rainbow, wherein the color channel index includes: brightness, contrast, and saturation; calculating the index score of the color channel index; and determining the color assessment result based on the index score and the color channel weight.

[0009] Furthermore, the quality assessment is an edge clarity assessment, and the rainbow information is quality assessed to obtain an assessment result, including: converting the processed rainbow information into grayscale rainbow information; convolving the Sobel operator and the grayscale rainbow information to obtain rainbow gradient information respectively, wherein the rainbow gradient information includes: rainbow horizontal gradient information and rainbow vertical gradient information; determining the gradient amplitude of the grayscale rainbow information based on the rainbow gradient information, scoring the edge clarity of the rainbow information based on the gradient amplitude to obtain a clarity score, and determining the clarity assessment result according to the clarity score and the edge clarity weight.

[0010] Furthermore, the quality assessment is a curve integrity assessment, and the quality assessment of the rainbow information is performed to obtain an assessment result, including: calculating the pixel points in the rainbow mask information through a polynomial curve fitting algorithm to obtain rainbow curve parameters, wherein the rainbow curve parameters include: length parameter and curvature parameter; determining the number of pixels and width information of the rainbow mask information according to the rainbow curve parameters; scoring the integrity of the rainbow curve based on the number of pixels and width information to obtain a curve score, and determining the curve integrity assessment result according to the curve score and curve weight information.

[0011] Furthermore, rainbow imaging effect feedback is formed according to the evaluation results, and the rainbow imaging is adjusted based on the rainbow imaging effect feedback, including: determining rainbow imaging parameters according to the rainbow imaging effect feedback, wherein the rainbow imaging parameters are rainbow information with the highest parameter quality; adjusting the nozzle parameters of the water mist nozzle according to the rainbow imaging parameters to adjust the rainbow imaging, wherein the nozzle parameters include: nozzle angle and nozzle pressure.

[0012] To achieve the above-mentioned purpose, according to the second aspect of the present application, a device for determining an artificial rainbow is provided. The device includes: an acquisition unit for acquiring an original image through an image acquisition device, wherein the original image carries rainbow information; a first determination unit for determining the rainbow information based on the original image and a target network framework, wherein the target network framework integrates: a target detection algorithm and a bilateral segmentation network, wherein the rainbow information includes: rainbow frame information and rainbow mask information; an evaluation unit for performing a quality evaluation on the rainbow information to obtain an evaluation result, wherein the quality evaluation includes at least one of the following: color channel index evaluation, edge clarity evaluation, and curve integrity evaluation; a second determination unit for generating rainbow imaging effect feedback based on the evaluation result, and adjusting the rainbow imaging based on the rainbow imaging effect feedback.

[0013] According to a third aspect of an embodiment of the present application, an electronic device is provided, comprising: a processor; and a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, any one of the above-mentioned methods for determining an artificial rainbow is implemented.

[0014] According to a fourth aspect of an embodiment of the present application, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the method for determining an artificial rainbow according to any one of the above items is implemented.

[0015] Through this application, the following steps are adopted: obtaining an original image through an image acquisition device, wherein the original image carries rainbow information; determining the rainbow information based on the original image and the target network framework, wherein the target network framework integrates: a target detection algorithm, a bilateral segmentation network, wherein the rainbow information includes: rainbow frame information, rainbow mask information; performing a quality assessment on the rainbow information to obtain an assessment result, wherein the quality assessment includes at least one of the following: color channel index assessment, edge clarity assessment, and curve integrity assessment; forming rainbow imaging effect feedback based on the assessment result, and adjusting the rainbow imaging based on the rainbow imaging effect feedback, thereby solving the problem in the related art that the rainbow image cannot be deployed and evaluated at the edge. By processing and evaluating the collected rainbow image in the edge device, and determining the target rainbow based on the assessment result, the effect of deploying and evaluating the rainbow image at the edge is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The accompanying drawings, which constitute part of this application, are intended to provide a further understanding of this application. The exemplary embodiments and descriptions of this application are intended to explain this application and do not constitute an improper limitation on this application. In the accompanying drawings:

[0017] Figure 1 is a flow chart of a method for determining an artificial rainbow according to an embodiment of the present application;

[0018] Figure 2 1 is a schematic diagram of the overall architecture of the method for determining an artificial rainbow provided in an embodiment of the present application;

[0019] Figure 3 1 is a schematic diagram of an instance segmentation CNN structure of a method for determining an artificial rainbow according to an embodiment of the present application;

[0020] Figure 4 Schematic diagram of the artificial rainbow BiSeNet network structure provided according to an embodiment of the present application;

[0021] Figure 5 is a schematic diagram of a device for determining an artificial rainbow according to an embodiment of the present application;

[0022] Figure 6 This is a schematic diagram of the network architecture of an electronic device provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0023] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in this application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0024] In order to enable those skilled in the art to better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this application.

[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] For ease of description, some nouns or terms involved in the embodiments of the present application are explained below:

[0027] YOLO (You Only Look Once, target detection algorithm): is an end-to-end real-time object detection model.

[0028] BiSeNet network: Bilateral Segmentation Network, is an effective semantic segmentation model.

[0029] The Sobel operator is an important processing method in the field of computer vision, mainly used to obtain the first-order gradient of digital images. It performs a convolution operation on the image to approximate the gradient value of the image brightness function, thereby achieving edge detection.

[0030] According to an embodiment of the present application, a method for determining an artificial rainbow is provided.

[0031] Figure 1 Flowchart of the method for determining an artificial rainbow according to an embodiment of the present application. Figure 1 As shown, the method includes the following steps:

[0032] Step S101: Acquire an original image through an image acquisition device, wherein the original image carries rainbow information.

[0033] Among them, the execution subject of this application can be an edge device, but the specific implementation process of this case is not limited to edge devices. The execution subject is also applicable to server devices, and this case does not make such a limitation.

[0034] This embodiment uses edge devices as examples of the execution subject. Edge devices can be Internet of Things (IoT) terminal devices, embedded systems (for example, Rockchip's embedded control platform), smart cameras and sensors. By deploying the image segmentation model at the edge, this application can significantly reduce data transmission latency, improve the response speed of the overall system, and fully tap the potential of existing hardware without relying on more powerful but power-hungry PC computing resources. In other words, this application achieves efficient distributed computing through image data processing on the edge device side, while being able to quickly respond to and process data requests, thereby improving the imaging efficiency of artificial rainbows.

[0035] Among them, the image acquisition device can use the Logitech C920 network HD camera to collect original images, realize the high-definition acquisition of original images by edge devices, and provide refined rainbow materials for subsequent image analysis.

[0036] Step S102 : determining rainbow information based on the original image and the target network framework, wherein the target network framework integrates: a target detection algorithm and a bilateral segmentation network, wherein the rainbow information includes: rainbow frame information and rainbow mask information.

[0037] The target network framework integrates a target detection algorithm and a bilateral segmentation network. The target detection algorithm can be the YOLO model algorithm, and the bilateral segmentation network can be the BiSeNet network. Rainbow information can be rainbow frame information or rainbow mask information. Both are obtained by algorithmically processing the original image to obtain rainbow frame information in the coordinate dimension and rainbow mask information in the feature point dimension, improving the accuracy of subsequent data processing of the original image.

[0038] In some embodiments, rainbow information is determined based on the original image and the target network framework, and is obtained by the following steps: inputting the original image into the target detection algorithm to obtain rainbow frame information; and using the rainbow frame information as the rainbow information.

[0039] For example, when the application receives a new raw image, it inputs the image into the YOLO algorithm model. The YOLO model processes the input raw image and quickly identifies rainbows in the image. For each rainbow detected, the model outputs one or more bounding boxes (equivalent to the rainbow frame information in this application), which include the rainbow's location and coordinate information. By determining the rainbow frame information, this application improves the accuracy of subsequent data processing of the raw image.

[0040] In some embodiments, rainbow information is determined based on the original image and the target network framework, which can be achieved through the following steps: obtaining feature map information of a preset number of layers of rainbow frame information in the target detection algorithm; inputting the feature map information into the bilateral segmentation network to obtain a rainbow segmented image; when it is recognized that the number of pixels in the target area in the rainbow segmentation image is greater than the preset number of pixels, the area with the largest number of pixels in the target area is used as rainbow mask information.

[0041] For example, Figure 2 As shown, the collected original image is input into the CNN convolutional neural network model combined with BiSeNet (in the YOLO model algorithm), and the mask information of the rainbow image is output. Based on this mask information, the imaging effect of the rainbow is evaluated. Specifically, the original image is input into Figure 3 In the CNN network shown in (in the YOLO model algorithm), the rainbow frame information is obtained; at the same time, the feature maps output by the 16th, 19th, and 22nd layers of the convolutional neural network CNN are used as Figure 4The spatial path of the BiSeNet shown in the figure is input and features are fused with the context path of the BiSeNet to output the mask information of the rainbow image. If there are multiple masks, and the number of pixels in the masked area (the area of ​​the recognized rainbow) in the rainbow segmentation image is greater than the preset number of pixels (i.e., the minimum rainbow area; if it is smaller than this area, the detected area is not considered a rainbow), the area with the largest number of pixels in the target area is used as the rainbow mask information. By determining the rainbow mask information, the interference of background noise and other irrelevant elements can be reduced, thereby improving the accuracy and efficiency of subsequent processing.

[0042] Specifically, Figure 3 This is an example diagram of the CNN structure for instance segmentation in this application, such as Figure 3 As shown in the figure, an original image with rainbow elements is described. YOLO combined with BiSeNet is used for instance segmentation task. Based on the Pytorch framework convolutional neural network model, the RGB color image is input and the instance segmentation result of the rainbow is output. After that, the segmentation result of the rainbow image after instance segmentation is quantitatively evaluated, thereby improving the imaging efficiency of artificial rainbow.

[0043] Step S103 , performing quality assessment on the rainbow information to obtain an assessment result, wherein the quality assessment includes at least one of the following: color channel index assessment, edge clarity assessment, and curve integrity assessment.

[0044] Specifically, the evaluation result can be a color evaluation result, a clarity evaluation result, or a curve integrity evaluation result, among which the color channel index evaluation of the rainbow information corresponds to the color evaluation result, the edge clarity evaluation of the rainbow information corresponds to the clarity evaluation result, and the curve integrity evaluation of the rainbow information corresponds to the clarity evaluation result; this application can improve data accuracy through the quality evaluation of rainbow information, and can also provide a deeper understanding of the characteristics and laws of rainbows, providing more accurate and reliable information support for research and applications in various fields.

[0045] In some embodiments, the quality assessment is a color channel index assessment, and the quality assessment of the rainbow information is performed to obtain an assessment result, which can be achieved by the following steps: processing the pixel area in the rainbow frame information except the rainbow mask information as black to obtain the processed rainbow information; performing spatial conversion on the processed rainbow information to output the color channel index of the rainbow, wherein the color channel index includes: brightness, contrast, and saturation; calculating the index score of the color channel index; and determining the color assessment result based on the index score and the color channel weight.

[0046] For example, a logical operation (such as an AND operation) is performed on the original image and the rainbow mask information, but in this scenario, the area outside the rainbow mask information is actually to be processed. Therefore, a black image of the same size as the original image is created, and then the white area within the rainbow mask is copied to this black image, keeping the area outside the mask black to obtain the processed rainbow information. The processed rainbow information is converted from a certain color space (such as RGB) to another space that is more convenient for analysis (such as HSV), and the color channel index of the rainbow is output. The index score of the color channel index is calculated. Afterwards, a weight can be assigned to each index based on the color channel index of the rainbow (brightness, contrast, saturation). This weight can be determined according to actual needs and application scenarios. For example, if we value the vividness of the rainbow more, we may give a higher weight to saturation. Similarly, the same logic applies to brightness and contrast, which will not be elaborated here. If the color evaluation result is 85 points (out of 100 points), this means that the rainbow color performance in the original image is quite good, with high brightness, moderate contrast and high saturation, making the rainbow very eye-catching and attractive to users in the picture. By extracting and analyzing the color channel indicators of the rainbow, this application can obtain a comprehensive evaluation result on the color performance of the rainbow. This evaluation result not only helps to understand the color characteristics of the rainbow itself, but also provides a valuable reference for the color evaluation of the entire photo.

[0047] In some embodiments, the quality assessment is an edge clarity assessment, and the quality assessment is performed on the rainbow information to obtain an assessment result, which can be achieved by the following steps: converting the processed rainbow information into grayscale rainbow information; convolving the Sobel operator and the grayscale rainbow information to obtain rainbow gradient information respectively, wherein the rainbow gradient information includes: rainbow horizontal gradient information and rainbow vertical gradient information; determining the gradient amplitude of the grayscale rainbow information based on the rainbow gradient information, scoring the edge clarity of the rainbow information based on the gradient amplitude to obtain a clarity score, and determining the clarity assessment result according to the clarity score and the edge clarity weight.

[0048] For example, as shown in Table 1, the Sobel operator (also referred to as the Sobel operator in this application) includes two 3x3 convolution kernels: one for calculating the gradient in the horizontal direction and the other for calculating the gradient in the vertical direction. The Sobel operator uses the grayscale changes of neighboring pixels to detect edges and has strong noise resistance.

[0049] Table 1

[0050]

[0051] For a rainbow image, these two Sobel operators can be used to calculate the gradient magnitude of the rainbow edge in the horizontal and vertical directions respectively. The specific steps are as follows:

[0052] ①Convert the rainbow image into grayscale rainbow information to simplify the calculation.

[0053] ② Convolve the horizontal and vertical Sobel operators with the grayscale image respectively to obtain the rainbow horizontal gradient information G X and rainbow vertical gradient information G Y .

[0054] ③The formula for calculating the gradient amplitude of each pixel is as follows:

[0055]

[0056] Among them, the larger the gradient amplitude, the more obvious the edge at that position.

[0057] Afterwards, the edge clarity of the rainbow information is scored based on the calculated gradient amplitude. A simple method is to set a threshold value, and the areas with gradient amplitudes higher than the threshold value are regarded as having clear edges, and the areas with gradient amplitudes lower than the threshold value are regarded as having blurred edges. These areas can then be further analyzed or classified. Gradient weight information is obtained. After evaluating the edge clarity, the gradient amplitude of each pixel value represents the clarity of the edge at the corresponding position. The larger the gradient amplitude, the higher the score, indicating the clearer the edge; the smaller the gradient amplitude, the blurred the edge, and the lower the score. By determining the edge clarity score, this application not only improves the image quality and visual effect, but also can provide strong support for subsequent image processing, feature extraction, recognition, segmentation and other tasks.

[0058] In some embodiments, the quality assessment is a curve integrity assessment, and the quality assessment of the rainbow information is performed to obtain an assessment result, which can be achieved by the following steps: calculating the pixel points in the rainbow mask information through a polynomial curve fitting algorithm to obtain rainbow curve parameters, wherein the rainbow curve parameters include: length parameter, curvature parameter; determining the number of pixels and width information of the rainbow mask information according to the rainbow curve parameters; scoring the integrity of the rainbow curve based on the number of pixels and width information to obtain a curve score, and determining the curve integrity assessment result according to the curve score and curve weight information.

[0059] For example, Figure 2 As shown in the figure, the detected pixel points are modeled using a polynomial curve fitting algorithm, and a smooth curve equation is fitted to describe the rainbow image. Specifically, by analyzing the parameters of the polynomial curve equation, the length parameter of the rainbow is calculated. rainbow And the maximum curvature parameter Curvature rainbow Among them, the length parameter and the maximum curvature parameter can be expressed by the following formula:

[0060]

[0061] Where y is the polynomial curve equation.

[0062] After that, the number of pixels Num of the connected area of ​​the rainbow mask information after instance segmentation is obtained by calling the opencv library pixels , then the width of the rainbow is Width rainbow It can be expressed as:

[0063]

[0064] Finally, for example, width accounts for 60% of the weight, length accounts for 20%, and curvature accounts for 20%. The total score of the two is calculated as the curve weight information, which reflects the degree of integrity of the rainbow curve. The curve weight information is used as the evaluation result. This weight information provides a valuable reference for subsequent image processing, feature extraction, recognition, and other tasks.

[0065] Step S104 : generating rainbow imaging effect feedback according to the evaluation result, and adjusting the rainbow imaging based on the rainbow imaging effect feedback.

[0066] Specifically, rainbow imaging effect feedback is formed according to the evaluation results, and rainbow imaging is adjusted based on the rainbow imaging effect feedback, which can be obtained through the following steps: rainbow imaging parameters are determined according to the rainbow imaging effect feedback, wherein the rainbow imaging parameters are the rainbow information with the highest parameter quality; nozzle parameters of the water mist nozzle are adjusted according to the rainbow imaging parameters to adjust the rainbow imaging, wherein the nozzle parameters include: nozzle angle and nozzle pressure.

[0067] Exemplarily, a fuzzy control algorithm is used to determine rainbow imaging effect feedback based on image analysis and evaluation results. Users can view the system's operating status in real time and manually adjust nozzle parameters. For example, based on the rainbow imaging parameters obtained from the rainbow imaging effect feedback, the angle and pressure of the water mist nozzle in the nozzle control module are adjusted in real time to achieve the optimal rainbow effect. For example, the nozzle angle adjustment range is set to 0-90 degrees and the pressure range is set to 0.1-0.5 MPa. Based on the evaluation results, this application obtains the target rainbow, improving the imaging effect of the artificial rainbow.

[0068] It should be noted that the fuzzy control algorithm is a control method that is very suitable for dealing with uncertainty and complex systems, especially when an accurate mathematical model is difficult to obtain. Fuzzy control is used to build a fuzzy rule base, infer the control strategy of the spraying system based on the input parameters, and dynamically adjust key factors such as water mist speed, spray angle and droplet size, so as to optimize the color, brightness and shape of the rainbow and achieve the ideal visual effect.

[0069] In summary, the method for determining an artificial rainbow provided by the embodiment of the present application obtains an original image through an image acquisition device, wherein the original image carries rainbow information; determines the rainbow information based on the original image and the target network framework, wherein the target network framework integrates: a target detection algorithm, a bilateral segmentation network, wherein the rainbow information includes: rainbow frame information, rainbow mask information; performs a quality assessment on the rainbow information to obtain an assessment result, wherein the quality assessment includes at least one of the following: color channel index assessment, edge clarity assessment, and curve integrity assessment; forms rainbow imaging effect feedback based on the assessment result, and adjusts the rainbow imaging based on the rainbow imaging effect feedback, thereby solving the problem in related technologies that rainbow images cannot be deployed and evaluated at the edge. By processing and evaluating the captured rainbow image in the edge device, and determining the target rainbow based on the assessment result, the effect of deploying and evaluating the rainbow image at the edge is achieved.

[0070] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0071] The present application also provides an artificial rainbow determination device. It should be noted that the artificial rainbow determination device of the present application can be used to execute the artificial rainbow determination method provided in the present application. The artificial rainbow determination device provided in the present application is described below.

[0072] Figure 5 FIG. 1 is a schematic diagram of a device for determining an artificial rainbow according to an embodiment of the present application. Figure 5 As shown, the device 500 includes: an acquisition unit 501 , a first determination unit 502 , an evaluation unit 503 , and a second determination unit 504 .

[0073] Specifically, the acquisition unit 501 is configured to acquire an original image through an image acquisition device, wherein the original image carries rainbow information;

[0074] A first determining unit 502 is configured to determine rainbow information based on the original image and a target network framework, wherein the target network framework integrates: a target detection algorithm and a bilateral segmentation network, wherein the rainbow information includes: rainbow frame information and rainbow mask information;

[0075] An evaluation unit 503 is configured to perform a quality evaluation on the rainbow information to obtain an evaluation result, wherein the quality evaluation includes at least one of the following: color channel index evaluation, edge clarity evaluation, and curve integrity evaluation;

[0076] The second determining unit 504 is configured to generate rainbow imaging effect feedback according to the evaluation result, and adjust the rainbow imaging based on the rainbow imaging effect feedback.

[0077] In summary, the artificial rainbow determination device provided by the embodiment of the present application obtains an original image through an image acquisition device via an acquisition unit 501, wherein the original image carries rainbow information; a first determination unit 502 determines the rainbow information based on the original image and a target network framework, wherein the target network framework integrates: a target detection algorithm and a bilateral segmentation network, wherein the rainbow information includes: rainbow frame information and rainbow mask information; an evaluation unit 503 performs a quality evaluation on the rainbow information to obtain an evaluation result, wherein the quality evaluation includes at least one of the following: color channel index evaluation, edge clarity evaluation, and curve integrity evaluation; a second determination unit 504 generates rainbow imaging effect feedback based on the evaluation result, and adjusts the rainbow imaging based on the rainbow imaging effect feedback, thereby solving the problem in related technologies that rainbow images cannot be deployed and evaluated at the edge. By processing and evaluating the collected rainbow image at the edge device, and determining the target rainbow based on the evaluation result, the effect of deploying and evaluating the rainbow image at the edge is achieved.

[0078] Optionally, in the artificial rainbow determination device provided in an embodiment of the present application, the first determination unit includes: a first input module, used to input the original image into the target detection algorithm to obtain rainbow frame information; and a first determination module, used to use the rainbow frame information as rainbow information.

[0079] Optionally, in the artificial rainbow determination device provided in an embodiment of the present application, the first determination unit includes: a cropping module, used to obtain feature map information of a preset number of layers of rainbow frame information in the target detection algorithm; a second input module, used to input the feature map information into a bilateral segmentation network to obtain a rainbow segmentation image; and an identification module, used to, when it is identified that the number of pixels in the target area in the rainbow segmentation image is greater than a preset number of pixels, use the area with the largest number of pixels in the target area as rainbow mask information.

[0080] Optionally, in the artificial rainbow determination device provided in the embodiment of the present application, the quality assessment is a color channel index assessment, and the assessment unit includes: a processing module, used to process the pixel area in the rainbow frame information except the rainbow mask information as black, to obtain processed rainbow information; a conversion module, used to perform spatial conversion on the processed rainbow information, and output the color channel index of the rainbow, wherein the color channel index includes: brightness, contrast, and saturation; a first calculation module, used to calculate the index score of the color channel index; and determine the color assessment result according to the index score and the color channel weight.

[0081] Optionally, in the artificial rainbow determination device provided in the embodiment of the present application, the quality assessment is an edge clarity assessment, and the assessment unit includes: a conversion module for converting the processed rainbow information into grayscale rainbow information; a convolution module for convolving the Sobel operator and the grayscale rainbow information to obtain rainbow gradient information respectively, wherein the rainbow gradient information includes: rainbow horizontal gradient information and rainbow vertical gradient information; a second determination module for determining the gradient amplitude of the grayscale rainbow information based on the rainbow gradient information, scoring the edge clarity of the rainbow information based on the gradient amplitude to obtain a clarity score, and determining a clarity assessment result according to the clarity score and the edge clarity weight.

[0082] Optionally, in the artificial rainbow determination device provided in the embodiment of the present application, the quality assessment is a curve integrity assessment, and the evaluation unit includes: a second calculation module, used to calculate the pixel points in the rainbow mask information through a polynomial curve fitting algorithm to obtain rainbow curve parameters, wherein the rainbow curve parameters include: length parameters, curvature parameters; a third determination module, used to determine the number of pixels and width information of the rainbow mask information according to the rainbow curve parameters; an evaluation module, used to score the integrity of the rainbow curve based on the number of pixels and width information to obtain a curve score, and determine the curve integrity assessment result according to the curve score and curve weight information.

[0083] Optionally, in the artificial rainbow determination device provided in the embodiment of the present application, the second determination unit includes: a fourth determination module, used to determine the rainbow imaging parameters based on the rainbow imaging effect feedback, wherein the rainbow imaging parameters are the rainbow information with the highest parameter quality; an adjustment module, used to adjust the nozzle parameters of the water mist nozzle according to the rainbow imaging parameters to adjust the rainbow imaging, wherein the nozzle parameters include: nozzle angle and nozzle pressure.

[0084] The device for determining an artificial rainbow includes a processor and a memory. The acquisition unit 501, the first determination unit 502, the evaluation unit 503, the second determination unit 504, etc. are all stored in the memory as program units, and the processor executes the program units stored in the memory to implement corresponding functions.

[0085] The processor includes a kernel, which retrieves the corresponding program unit from the memory. One or more kernels can be set, and the artificial rainbow is determined by adjusting the kernel parameters.

[0086] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0087] In an exemplary embodiment of the present application, a computer storage medium capable of implementing the above-mentioned method is also provided. A program product capable of implementing the above-mentioned method of the present specification is stored thereon. In some possible embodiments, various aspects of the present application can also be implemented in the form of a program product, which includes program code. When the program product is run on a terminal device, the program code is used to enable the terminal device to execute the steps according to various exemplary embodiments of the present application described in the above "Exemplary Method" section of this specification, such as the following steps: obtaining an original image through an image acquisition device, wherein the original image carries rainbow information; determining the rainbow information based on the original image and a target network framework, wherein the target network framework integrates: a target detection algorithm, a bilateral segmentation network, wherein the rainbow information includes: rainbow frame information, rainbow mask information; performing quality assessment on the rainbow information to obtain an assessment result, wherein the quality assessment includes at least one of the following: color channel index assessment, edge clarity assessment, and curve integrity assessment; generating rainbow imaging effect feedback based on the assessment result, and adjusting the rainbow imaging based on the rainbow imaging effect feedback.

[0088] In an optional implementation, the original image is input into a target detection algorithm to obtain rainbow frame information; and the rainbow frame information is used as the rainbow information.

[0089] In an optional embodiment: obtaining feature map information of a preset number of layers of rainbow frame information in a target detection algorithm; inputting the feature map information into a bilateral segmentation network to obtain a rainbow segmented image; and when it is recognized that the number of pixels in the target area in the rainbow segmented image is greater than a preset number of pixels, using the area with the largest number of pixels in the target area as rainbow mask information.

[0090] In an optional embodiment: the pixel area in the rainbow frame information excluding the rainbow mask information is processed as black to obtain processed rainbow information; the processed rainbow information is spatially converted to output rainbow color channel indicators, wherein the color channel indicators include: brightness, contrast, and saturation; the indicator score of the color channel indicator is calculated; and the color evaluation result is determined based on the indicator score and the color channel weight.

[0091] In an optional embodiment, the processed rainbow information is converted into grayscale rainbow information; the Sobel operator is convolved with the grayscale rainbow information to obtain rainbow gradient information, wherein the rainbow gradient information includes rainbow horizontal gradient information and rainbow vertical gradient information; the gradient amplitude of the grayscale rainbow information is determined based on the rainbow gradient information, the edge clarity of the rainbow information is scored based on the gradient amplitude to obtain a clarity score, and a clarity evaluation result is determined based on the clarity score and the edge clarity weight.

[0092] In an optional embodiment: pixels in the rainbow mask information are calculated using a polynomial curve fitting algorithm to obtain rainbow curve parameters, where the rainbow curve parameters include: a length parameter and a curvature parameter; the number of pixels and width information of the rainbow mask information are determined based on the rainbow curve parameters; the integrity of the rainbow curve is scored based on the number of pixels and the width information to obtain a curve score, and a curve integrity assessment result is determined based on the curve score and the curve weight information.

[0093] In an optional embodiment: rainbow imaging effect feedback is formed according to the evaluation results, and rainbow imaging is adjusted based on the rainbow imaging effect feedback, including: determining rainbow imaging parameters according to the rainbow imaging effect feedback, wherein the rainbow imaging parameters are rainbow information with the highest parameter quality; adjusting the nozzle parameters of the water mist nozzle according to the rainbow imaging parameters to adjust the rainbow imaging, wherein the nozzle parameters include: nozzle angle and nozzle pressure.

[0094] In an optional embodiment, the embodiments of the present application may further include a program product for implementing the above method, which may be a portable compact disc read-only memory (CD-ROM) and include program code, and can be run on a terminal device, such as a personal computer. However, the program product of the present application is not limited thereto. In this document, a readable storage medium may be any tangible medium containing or storing a program, which may be used by or in conjunction with an instruction execution system, apparatus, or device.

[0095] The program product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or component, or any combination thereof. More specific examples (a non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0096] A computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable signal medium may also be any readable medium other than a readable storage medium that can transmit, propagate, or transfer a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0097] The program code embodied on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0098] The program code for performing the operations of the present application can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, and the like, as well as conventional procedural programming languages ​​such as "C" or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, as a stand-alone software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0099] In addition, in an exemplary embodiment of the present application, an electronic device capable of implementing the above method is also provided.

[0100] Those skilled in the art will appreciate that various aspects of the present application can be implemented as systems, methods, or program products. Therefore, various aspects of the present application can be specifically implemented in the following forms: a complete hardware implementation, a complete software implementation (including firmware, microcode, etc.), or an implementation that combines hardware and software aspects, which may be collectively referred to herein as a "circuit," "module," or "system."

[0101] Refer to the following Figure 6 hereinafter, an electronic device 600 according to this embodiment of the present application is described. Figure 6 The electronic device 600 shown is merely an example and should not limit the functions and scope of use of the embodiments of the present application.

[0102] like Figure 6 As shown, electronic device 600 is implemented as a general-purpose computing device. Components of electronic device 600 may include, but are not limited to, the aforementioned at least one processing unit 610, the aforementioned at least one storage unit 620, a bus 630 connecting various system components (including storage unit 620 and processing unit 610), and a display unit 640.

[0103] The storage unit stores program code, and the program code can be executed by the processing unit 610, so that the processing unit 610 performs the steps described in the "Exemplary Method" section of the present specification according to various exemplary embodiments of the present application. For example, the processing unit 610 can perform the following steps: obtaining an original image through an image acquisition device, wherein the original image carries rainbow information; determining the rainbow information based on the original image and a target network framework, wherein the target network framework integrates: a target detection algorithm, a bilateral segmentation network, wherein the rainbow information includes: rainbow frame information, rainbow mask information; performing quality evaluation on the rainbow information to obtain an evaluation result, wherein the quality evaluation includes at least one of the following: color channel index evaluation, edge clarity evaluation, and curve integrity evaluation; forming rainbow imaging effect feedback based on the evaluation result, and adjusting the rainbow imaging based on the rainbow imaging effect feedback.

[0104] In an optional implementation, the original image is input into a target detection algorithm to obtain rainbow frame information; and the rainbow frame information is used as the rainbow information.

[0105] In an optional embodiment: obtaining feature map information of a preset number of layers of rainbow frame information in a target detection algorithm; inputting the feature map information into a bilateral segmentation network to obtain a rainbow segmented image; and when it is recognized that the number of pixels in the target area in the rainbow segmented image is greater than a preset number of pixels, using the area with the largest number of pixels in the target area as rainbow mask information.

[0106] In an optional embodiment: the pixel area in the rainbow frame information excluding the rainbow mask information is processed as black to obtain processed rainbow information; the processed rainbow information is spatially converted to output rainbow color channel indicators, wherein the color channel indicators include: brightness, contrast, and saturation; the indicator score of the color channel indicator is calculated; and the color evaluation result is determined based on the indicator score and the color channel weight.

[0107] In an optional embodiment, the processed rainbow information is converted into grayscale rainbow information; the Sobel operator is convolved with the grayscale rainbow information to obtain rainbow gradient information, wherein the rainbow gradient information includes rainbow horizontal gradient information and rainbow vertical gradient information; the gradient amplitude of the grayscale rainbow information is determined based on the rainbow gradient information, the edge clarity of the rainbow information is scored based on the gradient amplitude to obtain a clarity score, and a clarity evaluation result is determined based on the clarity score and the edge clarity weight.

[0108] In an optional embodiment: pixels in the rainbow mask information are calculated using a polynomial curve fitting algorithm to obtain rainbow curve parameters, where the rainbow curve parameters include: a length parameter and a curvature parameter; the number of pixels and width information of the rainbow mask information are determined based on the rainbow curve parameters; the integrity of the rainbow curve is scored based on the number of pixels and the width information to obtain a curve score, and a curve integrity assessment result is determined based on the curve score and the curve weight information.

[0109] In an optional embodiment: rainbow imaging effect feedback is formed according to the evaluation results, and rainbow imaging is adjusted based on the rainbow imaging effect feedback, including: determining rainbow imaging parameters according to the rainbow imaging effect feedback, wherein the rainbow imaging parameters are rainbow information with the highest parameter quality; adjusting the nozzle parameters of the water mist nozzle according to the rainbow imaging parameters to adjust the rainbow imaging, wherein the nozzle parameters include: nozzle angle and nozzle pressure.

[0110] The storage unit 620 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 6201 and / or a cache memory unit 6202 , and may further include a read-only memory unit (ROM) 6203 .

[0111] The storage unit 620 may also include a program / utility 6204 having a set (at least one) of program modules 6205, such program modules 6205 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0112] Bus 630 may represent one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0113] The electronic device 600 can also communicate with one or more external devices 700 (e.g., a keyboard, a pointing device, a Bluetooth device, etc.), one or more devices that enable a user to interact with the electronic device 600, and / or any device that enables the electronic device 600 to communicate with one or more other computing devices (e.g., a router, a modem, etc.). Such communication can occur via an input / output (I / O) interface 650. Furthermore, the electronic device 600 can communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network such as the Internet) via a network adapter 660. As shown, the network adapter 660 communicates with other modules of the electronic device 600 via a bus 630. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in conjunction with the electronic device 600, including but not limited to microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0114] Through the description of the above embodiments, it is easy for those skilled in the art to understand that the example embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solution according to the embodiments of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the embodiments of the present application.

[0115] Furthermore, the above-mentioned figures are merely illustrative of the processes included in the methods according to exemplary embodiments of the present application and are not intended to be limiting. It is readily understood that the processes illustrated in the above-mentioned figures do not indicate or limit the temporal order of these processes. Furthermore, it is readily understood that these processes may be executed synchronously or asynchronously, for example, in multiple modules.

[0116] Other embodiments of the present invention will readily occur to those skilled in the art after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present invention that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, and the true scope and spirit of the present application are indicated by the claims.

Claims

1. A method for determining an artificial rainbow, characterized in that: Applied to edge devices, including: Acquire an original image through an image acquisition device, wherein the original image carries rainbow information; Determining rainbow information based on the original image and a target network framework, wherein the target network framework integrates: a target detection algorithm and a bilateral segmentation network, wherein the rainbow information includes: rainbow frame information and rainbow mask information; Performing a quality assessment on the rainbow information to obtain an assessment result, wherein the quality assessment includes at least one of the following: color channel index assessment, edge clarity assessment, and curve integrity assessment; forming rainbow imaging effect feedback according to the evaluation result, and adjusting rainbow imaging based on the rainbow imaging effect feedback; Determining rainbow information based on the original image and the target network framework includes: Inputting the original image into the target detection algorithm to obtain the rainbow frame information; Using the rainbow frame information as the rainbow information; Determining rainbow information based on the original image and the target network framework also includes: Obtaining feature map information of a preset number of layers of the rainbow frame information in the target detection algorithm; Inputting the feature map information into the bilateral segmentation network to obtain a rainbow segmented image; When it is identified that the number of pixel points of the target area in the rainbow segmented image is greater than the preset number of pixel points, the area with the largest number of pixel points in the target area is used as the rainbow mask information.

2. The method according to claim 1, characterized in that The quality assessment is the color channel index assessment, and the quality assessment of the rainbow information is performed to obtain an assessment result, including: Processing the pixel area in the rainbow frame information except the rainbow mask information to black, to obtain processed rainbow information; Performing spatial conversion on the processed rainbow information to output rainbow color channel indicators, wherein the color channel indicators include: brightness, contrast, and saturation; Calculating an indicator score of the color channel indicator; A color evaluation result is determined based on the indicator score and the color channel weight.

3. The method according to claim 2, characterized in that The quality assessment is the edge clarity assessment, and the rainbow information is subjected to quality assessment to obtain an assessment result, including: converting the processed rainbow information into grayscale rainbow information; Convolving the Sobel operator with the grayscale rainbow information to obtain rainbow gradient information, wherein the rainbow gradient information includes: rainbow horizontal gradient information and rainbow vertical gradient information; The gradient amplitude of the grayscale rainbow information is determined based on the rainbow gradient information, the edge clarity of the rainbow information is scored based on the gradient amplitude to obtain a clarity score, and a clarity evaluation result is determined according to the clarity score and an edge clarity weight.

4. The method according to claim 1, wherein The quality assessment is the curve integrity assessment, and the rainbow information is subjected to quality assessment to obtain an assessment result, including: Calculating the pixel points in the rainbow mask information by a polynomial curve fitting algorithm to obtain rainbow curve parameters, wherein the rainbow curve parameters include: a length parameter and a curvature parameter; Determine the number of pixels and width information of the rainbow mask information according to the rainbow curve parameters; The integrity of the rainbow curve is scored based on the number of pixels and the width information to obtain a curve score, and a curve integrity evaluation result is determined according to the curve score and the curve weight information.

5. The method according to claim 1, wherein Generating rainbow imaging effect feedback according to the evaluation result, and adjusting rainbow imaging based on the rainbow imaging effect feedback, including: Determining rainbow imaging parameters according to the rainbow imaging effect feedback, wherein the rainbow imaging parameters are rainbow information with the highest parameter quality; The nozzle parameters of the water mist nozzle are adjusted according to the rainbow imaging parameters to adjust the rainbow imaging, wherein the nozzle parameters include: nozzle angle and nozzle pressure.

6. A device for determining an artificial rainbow, characterized in that: Applied to edge devices, including: an acquisition unit, configured to acquire an original image through an image acquisition device, wherein the original image carries rainbow information; A first determining unit is configured to determine rainbow information based on the original image and a target network framework, wherein the target network framework integrates: a target detection algorithm and a bilateral segmentation network, wherein the rainbow information includes: rainbow frame information and rainbow mask information; An evaluation unit, configured to perform a quality evaluation on the rainbow information to obtain an evaluation result, wherein the quality evaluation includes at least one of the following: color channel index evaluation, edge clarity evaluation, and curve integrity evaluation; a second determining unit, configured to determine a target rainbow according to the evaluation result; A first determining unit is configured to: input the original image into the target detection algorithm to obtain the rainbow frame information; Using the rainbow frame information as the rainbow information; The first determining unit is further configured to: Obtaining feature map information of a preset number of layers of the rainbow frame information in the target detection algorithm; Inputting the feature map information into the bilateral segmentation network to obtain a rainbow segmented image; When it is identified that the number of pixel points of the target area in the rainbow segmented image is greater than the preset number of pixel points, the area with the largest number of pixel points in the target area is used as the rainbow mask information.

7. A computer-readable storage medium, characterized in that The storage medium includes a stored program, wherein the program executes the method for determining an artificial rainbow according to any one of claims 1 to 5.

8. An electronic device, characterized in that: include: One or more processors, a memory, a display device, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include a method for executing the artificial rainbow determination method according to any one of claims 1 to 5.

Images