High-resolution large-breadth optical image intelligent stripe detection method and device
By dividing image blocks and matching histogram to determine the target area and inputting it into the detection model, the problem of fringe noise interference in high-score large-scale wide images is solved, and the detection efficiency and effect are improved.
Patent Information
- Application Number
- CN202410934334.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-12
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2044-07-12
AI Technical Summary
There is fringe noise interference in high-score large-scale wide images, resulting in a decrease in image quality and availability. The existing detection methods are inefficient and have poor results.
By dividing the image to be detected into initial image blocks, the candidate image blocks are determined and a grayscale histogram is generated, the histograms of the candidate and target area images are matched to determine the target area, and input to the stripe detection and quality evaluation model to obtain the detection results.
The data processing volume is reduced, the stripe detection efficiency and effect are improved, and the quality of high-score large-scale wide images are improved.
Smart Images

Figure CN119027368A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the field of artificial intelligence technology, and in particular to a method and device for intelligent stripe detection of high-resolution and wide-width optical images. Background Art
[0002] With the continuous development of high-altitude optical camera imaging technology, more and more wide-width and high-resolution image data are obtained. Due to the special properties of high-resolution and wide-width images, that is, the large imaging area, the probability of image quality being affected by cloud and fog obstruction, atmospheric interference, lighting conditions, etc. is also increasing, resulting in more stripe noise in the image. Stripe noise is a common interference, which is usually caused by factors such as uneven sensor scanning lines and sensor channel responses. Stripe interference in the image will cause discontinuity in image brightness, color and texture, thereby affecting the quality and availability of high-resolution and wide-width images.
[0003] In the related art, there are problems such as poor stripe detection effect and low detection timeliness in high-resolution and wide-width image detection. Summary of the invention
[0004] The present disclosure aims to solve one of the technical problems in the related art at least to some extent.
[0005] To this end, the purpose of the present invention is to propose a method, device, electronic device and storage medium for intelligent stripe detection of high-resolution and wide-width optical images, which can determine the target area image from the high-resolution and wide-width image to be detected with a large amount of data, thereby reducing the subsequent data processing volume and improving the stripe detection efficiency. Moreover, since the processing complexity of the target area image is relatively low, the stripe detection effect of the high-resolution and wide-width image can be improved.
[0006] To achieve the above-mentioned purpose, the first aspect of the present disclosure provides a high-resolution and wide-width optical image intelligent stripe detection method, including:
[0007] Acquire a high-resolution and large-width image to be detected, wherein the high-resolution and large-width image to be detected includes: a plurality of initial image blocks;
[0008] Determine at least one candidate image block from a plurality of initial image blocks;
[0009] Determine a first grayscale histogram corresponding to each candidate image block, and determine a second grayscale histogram corresponding to the high-resolution and wide-width image to be detected based on the first grayscale histogram corresponding to each candidate image block;
[0010] Determine a target area image from the high-resolution and wide-width image to be detected, and determine a third grayscale histogram corresponding to the target area image;
[0011] Determining a degree of matching between the third grayscale histogram and the second grayscale histogram;
[0012] When it is determined that the matching degree is greater than or equal to the matching degree threshold, the target area image is respectively input into the stripe detection model and the quality evaluation model to obtain the stripe detection result output by the stripe detection model and the quality evaluation result output by the quality evaluation model.
[0013] To achieve the above-mentioned purpose, the second aspect of the present disclosure provides a high-resolution and wide-width optical image intelligent stripe detection device, comprising:
[0014] An acquisition module is used to acquire a high-resolution and large-width image to be detected, wherein the high-resolution and large-width image to be detected includes: a plurality of initial image blocks;
[0015] A first determination module, configured to determine at least one candidate image block from a plurality of initial image blocks;
[0016] A second determination module is used to determine a first grayscale histogram corresponding to each candidate image block, and determine a second grayscale histogram corresponding to the high-resolution and wide-width image to be detected based on the first grayscale histogram corresponding to each candidate image block;
[0017] A third determination module is used to determine a target area image from the high-resolution and wide-width image to be detected, and to determine a third grayscale histogram corresponding to the target area image;
[0018] A fourth determination module, used to determine a matching degree between the third grayscale histogram and the second grayscale histogram;
[0019] The processing module is used to input the target area image into the stripe detection model and the quality evaluation model respectively when determining that the matching degree is greater than or equal to the matching degree threshold, so as to obtain the stripe detection result output by the stripe detection model and the quality evaluation result output by the quality evaluation model.
[0020] The electronic device proposed in the third aspect embodiment of the present disclosure includes: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the high-resolution and wide-width optical image intelligent stripe detection method proposed in the first aspect embodiment of the present disclosure is implemented.
[0021] The fourth aspect embodiment of the present disclosure proposes a non-temporary computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the high-resolution and wide-width optical image intelligent stripe detection method proposed in the first aspect embodiment of the present disclosure is implemented.
[0022] The fifth aspect embodiment of the present disclosure proposes a computer program product. When the instructions in the computer program product are executed by a processor, the high-resolution and large-width optical image intelligent stripe detection method proposed in the first aspect embodiment of the present disclosure is executed.
[0023] The high-resolution and large-width optical image intelligent stripe detection method, device, electronic device and storage medium provided by the present disclosure at least include the following beneficial effects: obtaining a high-resolution and large-width image to be detected, wherein the high-resolution and large-width image to be detected includes: a plurality of initial image blocks, then determining at least one candidate image block from the plurality of initial image blocks, then determining a first grayscale histogram corresponding to each candidate image block, and determining a second grayscale histogram corresponding to the high-resolution and large-width image to be detected based on the first grayscale histogram corresponding to each candidate image block, then determining a target area image from the high-resolution and large-width image to be detected, and determining a third grayscale histogram corresponding to the target area image; The target area image is input into the stripe detection model and the quality evaluation model respectively to obtain the stripe detection result output by the stripe detection model and the quality evaluation result output by the quality evaluation model. Thus, the target area image can be determined from the high-resolution and large-width image to be detected with a large amount of data, thereby reducing the subsequent data processing amount and improving the stripe detection efficiency. Moreover, since the processing complexity of the target area image is low, the stripe detection effect of the high-resolution and large-width image can be improved.
[0024] Additional aspects and advantages of the present disclosure will be given in part in the following description and in part will be obvious from the following description or learned through practice of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS
[0025] The above and / or additional aspects and advantages of the present disclosure will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:
[0026] Figure 1 It is a flowchart of a method for intelligent stripe detection of high-resolution and wide-width optical images proposed in one embodiment of the present disclosure;
[0027] Figure 2A is a structural schematic diagram of a stripe detection model proposed in an embodiment of the present disclosure;
[0028] Figure 2B is a schematic diagram of the model structure of a quality evaluation model proposed in an embodiment of the present disclosure;
[0029] Figure 3 is a flow chart of a method for intelligent stripe detection of high-resolution and wide-width optical images proposed in another embodiment of the present disclosure;
[0030] Figure 4 It is a structural schematic diagram of a high-resolution and wide-width optical image intelligent stripe detection device proposed in an embodiment of the present disclosure;
[0031] Figure 5 A block diagram of an exemplary electronic device suitable for implementing embodiments of the present disclosure is shown. DETAILED DESCRIPTION
[0032] Embodiments of the present disclosure are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present disclosure, and are not to be construed as limitations of the present disclosure. On the contrary, the embodiments of the present disclosure include all changes, modifications, and equivalents that fall within the spirit and connotation of the appended claims.
[0033] Figure 1 It is a flow chart of a method for intelligent stripe detection of high-resolution and wide-width optical images proposed in one embodiment of the present disclosure.
[0034] Among them, it should be noted that the executor of the high-resolution and large-width optical image intelligent stripe detection method of this embodiment is a high-resolution and large-width optical image intelligent stripe detection device, which can be implemented by software and / or hardware. The device can be configured in an electronic device, and the electronic device can include but is not limited to a terminal, a server, etc. For example, the terminal can be a mobile phone, a handheld computer, etc.
[0035] like Figure 1 As shown, the high-resolution and wide-width optical image intelligent stripe detection method includes:
[0036] S101: Acquire a high-resolution and large-width image to be detected, wherein the high-resolution and large-width image to be detected includes: a plurality of initial image blocks.
[0037] Among them, the high-resolution and large-width images have a large amount of image data, high image resolution, and a large imaging area. The high-resolution and large-width images can be, for example: optical remote sensing images, images with geographic information, etc., without limitation.
[0038] Among them, stripe noise is a common image interference. Stripe noise is usually caused by factors such as uneven sensor scanning lines and sensor channel responses. Stripe noise in the image can cause discontinuities in image brightness, color and texture, thereby affecting image quality.
[0039] The high-resolution and large-width image to be detected refers to the high-resolution and large-width image on which stripe detection is currently to be performed.
[0040] In the embodiment of the present disclosure, the high-resolution and large-width image to be detected can be obtained by acquiring the high-resolution and large-width image to be detected based on a preset area by a remote sensing image acquisition device, or any remote sensing image can be obtained from a remote sensing image database storing multiple remote sensing images as the high-resolution and large-width image to be detected, and there is no limitation on this.
[0041] The high-resolution and large-width image to be detected includes: a plurality of initial image blocks, and the plurality of initial image blocks are obtained by performing rasterization processing on the high-resolution and large-width image to be detected.
[0042] That is to say, in the embodiment of the present disclosure, the high-resolution and large-width image to be detected may be rasterized based on a preset image block size to obtain a plurality of initial image blocks of the same size.
[0043] S102: Determine at least one candidate image block from a plurality of initial image blocks.
[0044] In the embodiment of the present disclosure, after obtaining the high-resolution and large-width image to be detected, at least one candidate image block may be determined from a plurality of initial image blocks.
[0045] It can be understood that after the high-resolution and wide-width image to be detected is divided into multiple initial image blocks, some of the initial image blocks may not contain semantic information. Therefore, to determine at least one candidate image block from the multiple initial image blocks, the initial image blocks that do not contain semantic information from the multiple initial image blocks can be deleted, and the remaining initial image blocks are used as candidate image blocks.
[0046] Optionally, in some embodiments, determining at least one candidate image block from multiple initial image blocks may be by determining a second number of second pixels in each initial image block, and when the second number is greater than or equal to a number threshold, using the initial image block corresponding to the second number as a candidate image block.
[0047] The number of pixels in each initial image block is the second number.
[0048] That is to say, in the embodiment of the present disclosure, after obtaining the high-resolution and large-width image to be detected, the second number of second pixel points in each initial image block in the high-resolution and large-width image to be detected can be determined, and when it is determined that the second number is greater than or equal to the number threshold, the initial image block corresponding to the second number is used as a candidate image block.
[0049] For example, when it is determined that the second number of second pixels in the initial image block is less than 5000*5000, the initial image block is deleted and the remaining initial image blocks are used as candidate image blocks.
[0050] S103: Determine a first grayscale histogram corresponding to each candidate image block, and determine a second grayscale histogram corresponding to the high-resolution and wide-width image to be detected based on the first grayscale histogram corresponding to each candidate image block.
[0051] Among them, the grayscale histogram corresponding to the candidate image block is the first grayscale histogram, and the grayscale histogram corresponding to the high-resolution and large-width image to be detected is the second grayscale histogram.
[0052] That is, in the embodiment of the present disclosure, after determining at least one candidate image block from multiple initial image blocks, grayscale histogram statistics can be performed on each candidate image block to obtain a first grayscale histogram corresponding to each candidate image block.
[0053] In the disclosed embodiment, after determining the first grayscale histogram corresponding to each candidate image block, a second grayscale histogram corresponding to the high-resolution and wide-width image to be detected can be determined based on the first grayscale histogram corresponding to each candidate image block.
[0054] Optionally, in some embodiments, based on the first grayscale histogram corresponding to each candidate image block, a second grayscale histogram corresponding to the high-resolution and large-width image to be detected is determined. The quantitative average of the first quantity corresponding to each grayscale level in multiple first grayscale histograms is determined, and then the second grayscale histogram corresponding to the high-resolution and large-width image to be detected is generated based on the quantitative average of the first pixels corresponding to each grayscale level.
[0055] The first grayscale histogram includes: a first number of first pixel points corresponding to each grayscale level.
[0056] In the first grayscale histogram, the pixel point corresponding to each grayscale level is the first pixel point, and the number of the first pixel points is the first number.
[0057] That is to say, in the embodiment of the present disclosure, the quantitative average of the first quantity corresponding to each gray level in multiple first gray histograms can be determined, and then a second gray histogram corresponding to the high-resolution and wide-width image to be detected can be generated based on the quantitative average of the first pixels corresponding to each gray level.
[0058] S104: Determine a target area image from the high-resolution and wide-width image to be detected, and determine a third grayscale histogram corresponding to the target area image.
[0059] The local image for subsequent stripe detection determined from the high-resolution and large-width image to be detected is a target area image, and the target area image has typical ground feature of the high-resolution and large-width image to be detected.
[0060] The grayscale histogram corresponding to the target area image is the third grayscale histogram.
[0061] In some embodiments, determining the target area image from the high-resolution and large-width image to be detected may be to determine a local image located at the center of the image from the high-resolution and large-width image to be detected as the target area image based on a preset image size.
[0062] In other embodiments, the target area image can be determined from the high-resolution and large-width image to be detected, and the high-resolution and large-width image to be detected can also be processed based on a pre-trained image segmentation model. That is, the high-resolution and large-width image to be detected can be input into a pre-trained image segmentation model to obtain the target area image output by the image segmentation model. There is no limitation to this.
[0063] In the disclosed embodiment, the size of the high-resolution and large-width image to be detected is h*w.
[0064] Optionally, in some embodiments, the target area image is determined from the high-resolution and large-width image to be detected, and a local image with an image size of x*w located at the center of the image can be determined from the high-resolution and large-width image to be detected as the target area image, where x <h。
[0065] Among them, after a large number of experiments, it was found that the size of typical ground objects usually accounts for 1 / 4 of the full image, and the location is usually in the middle of the panoramic image. Because image stripes usually run through the entire panoramic image, if there are stripes in the middle part, a panoramic image must have stripes.
[0066] For example, according to the mechanism that image stripe noise permeates the entire image, the middle h / 2 rows may be selected as the starting row of the target area image, and then a data block with an image size of x*w (1000*w) may be determined as the target area image.
[0067] S105: Determine the matching degree between the third grayscale histogram and the second grayscale histogram.
[0068] The matching degree between the third grayscale histogram and the second grayscale histogram may refer to the similarity between the two grayscale histograms (eg, histogram intersection, chi-square distance, Bhattacharyya distance, etc.), and there is no limitation to this.
[0069] That is to say, in the embodiments of the present disclosure, the Bhattacharyya distance or the chi-square distance between the second grayscale histogram and the third grayscale histogram may be determined, and the Bhattacharyya distance or the chi-square distance between the second grayscale histogram and the third grayscale histogram determined above may be used as the degree of matching between the second grayscale histogram and the third grayscale histogram, and there is no limitation on this.
[0070] S106: When it is determined that the matching degree is greater than or equal to the matching degree threshold, the target area image is respectively input into the stripe detection model and the quality evaluation model to obtain the stripe detection result output by the stripe detection model and the quality evaluation result output by the quality evaluation model.
[0071] In the embodiments of the present disclosure, see Figure 2A , Figure 2A is a schematic diagram of the structure of a stripe detection model proposed in an embodiment of the present disclosure, such as Figure 2A As shown, the model structure of the stripe detection model includes in sequence: a first standard convolution module CBL, a first backbone network, a channel-based lightweight re-encoding module CLR, a second CBL and a second backbone network.
[0072] In the embodiments of the present disclosure, see Figure 2B , Figure 2B is a schematic diagram of the model structure of the quality evaluation model proposed in one embodiment of the present disclosure, such as Figure 2B As shown, the model structure of the quality evaluation model includes: a third CBL, a third backbone network, a fourth backbone network and a fourth CBL in sequence.
[0073] The CBL module contains convolutional layers, batch normalization layers, and leaky activation layers. The convolutional layer is used for feature extraction, and the regularization layer helps the stability of the network, so that the model can converge quickly during training. The leaky activation function is as follows:
[0074] f(x)=a*xx<0;
[0075] f(x)=xx≥0;
[0076] In the disclosed embodiment, the backbone network of the quality assessment model adopts a residual structure.
[0077] In the disclosed embodiment, a residual network structure is used for lightweight model design, a CBL (convolution, batch regularization, leaky activation function) module is used instead of a traditional convolution layer to extract features of different scales, and a CLR module is used to fuse features of different features. This allows the stripe detection neural network to have smaller network parameters and a faster detection speed without reducing the feature expression capability of the neural network.
[0078] In the disclosed embodiment, the stripe detection model is pre-trained based on a first mapping relationship between a sample high-resolution and large-width image and annotated stripe detection results corresponding to the sample high-resolution and large-width image, and the quality evaluation model is pre-trained based on a second mapping relationship between a sample high-resolution and large-width image and annotated signal-to-noise ratio corresponding to the sample high-resolution and large-width image.
[0079] That is to say, in the embodiments of the present disclosure, an initial stripe detection model may be iteratively trained based on the sample high-resolution and wide-width images, and the annotated stripe detection results corresponding to the sample high-resolution and wide-width images, until the initial stripe detection model converges to obtain a stripe detection model; an initial quality evaluation model may be iteratively trained based on the sample high-resolution and wide-width images, and the annotated signal-to-noise ratio corresponding to the sample high-resolution and wide-width images, until the initial stripe detection model converges to obtain a quality evaluation model; and then, a subsequent high-resolution and wide-width optical image intelligent stripe detection method may be executed based on the trained stripe detection model and quality evaluation model. For details, please refer to the subsequent embodiments.
[0080] The stripe detection result is 0 or 1, where 0 indicates that there is no stripe noise in the high-resolution and large-width image to be detected, and 1 indicates that there is stripe noise in the high-resolution and large-width image to be detected.
[0081] The quality evaluation result may be used to quantitatively evaluate the image quality of the high-resolution and wide-width image to be detected. The quality evaluation result may be, for example, an image signal-to-noise ratio, which is not limited.
[0082] In the embodiment of the present disclosure, after determining the degree of matching between the third grayscale histogram and the second grayscale histogram, the degree of matching can be compared with a predetermined matching degree threshold, and when the degree of matching is greater than or equal to the matching degree threshold, the target area image is respectively input into the stripe detection model and the quality evaluation model to obtain the stripe detection result output by the stripe detection model and the quality evaluation result output by the quality evaluation model.
[0083] In the disclosed embodiment, a high-resolution and large-width image to be detected is obtained, wherein the high-resolution and large-width image to be detected includes: a plurality of initial image blocks, and then at least one candidate image block is determined from the plurality of initial image blocks, and then a first grayscale histogram corresponding to each candidate image block is determined, and then a second grayscale histogram corresponding to the high-resolution and large-width image to be detected is determined according to the first grayscale histogram corresponding to each candidate image block, and then a target area image is determined from the high-resolution and large-width image to be detected, and a third grayscale histogram corresponding to the target area image is determined, and then a matching degree between the third grayscale histogram and the second grayscale histogram is determined, and when it is determined that the matching degree is greater than or equal to a matching degree threshold, the target area image is respectively input into a stripe detection model and a quality evaluation model to obtain a stripe detection result output by the stripe detection model, and a quality evaluation result output by the quality evaluation model, thereby, the target area image can be determined from the high-resolution and large-width image to be detected with a large amount of data, so that the subsequent data processing amount can be reduced, thereby improving the stripe detection efficiency, and because the processing complexity of the target area image is low, the stripe detection effect of the high-resolution and large-width image can be improved.
[0084] FIG2 is a flow chart of a method for intelligent stripe detection in high-resolution and wide-width optical images according to another embodiment of the present disclosure.
[0085] As shown in FIG2 , the high-resolution and wide-width optical image intelligent stripe detection method includes:
[0086] S301: Acquire a high-resolution and large-width image to be detected, wherein the high-resolution and large-width image to be detected includes: a plurality of initial image blocks.
[0087] S302: Determine at least one candidate image block from a plurality of initial image blocks.
[0088] S303: Determine a first grayscale histogram corresponding to each candidate image block, and determine a second grayscale histogram corresponding to the high-resolution and wide-width image to be detected based on the first grayscale histogram corresponding to each candidate image block.
[0089] S304: Determine a target area image from the high-resolution and wide-width image to be detected, and determine a third grayscale histogram corresponding to the target area image.
[0090] S305: Determine the matching degree between the third grayscale histogram and the second grayscale histogram.
[0091] S306: When it is determined that the matching degree is greater than or equal to the matching degree threshold, the target area image is respectively input into the stripe detection model and the quality evaluation model to obtain the stripe detection result output by the stripe detection model and the quality evaluation result output by the quality evaluation model.
[0092] The specific description of S301 - S306 can be found in the above embodiment, which will not be repeated here.
[0093] S307: When the matching degree is less than the matching degree threshold, return to the operation of determining the target area image from the high-resolution and large-width image to be detected, wherein in the operation of returning to the operation of determining the target area image from the high-resolution and large-width image to be detected, the image size of the target area image is y*w, x <y。
[0094] In the embodiment of the present disclosure, after determining the degree of matching between the third grayscale histogram and the second grayscale histogram, the degree of matching can be compared with a predetermined matching degree threshold, and when the degree of matching is less than the matching degree threshold, the operation of determining the target area image from the high-resolution and wide-width image to be detected is returned.
[0095] In an embodiment of the present disclosure, when returning to perform the operation of determining the target region image from the high-resolution and wide-field image to be detected, the target region image with a larger size (y*w, x < y) in the initial operation can be selected. That is, in the h dimension direction of the target region image determined in the initial operation, 500 pixels can be extended in each direction to form a new target region image of 3000*w pixels.
[0096] In an embodiment of the present disclosure, by obtaining a high-resolution and wide-field image to be detected, where the high-resolution and wide-field image to be detected includes: a plurality of initial image blocks, then determining at least one candidate image block from the plurality of initial image blocks, then determining a first grayscale histogram corresponding to each candidate image block, and according to the first grayscale histogram corresponding to each candidate image block, determining a second grayscale histogram corresponding to the high-resolution and wide-field image to be detected, then determining a target region image from the high-resolution and wide-field image to be detected, and determining a third grayscale histogram corresponding to the target region image, then determining the matching degree between the third grayscale histogram and the second grayscale histogram. When it is determined that the matching degree is greater than or equal to the matching degree threshold, the target region image is respectively input into the stripe detection model and the quality evaluation model to obtain the stripe detection result output by the stripe detection model and the quality evaluation result output by the quality evaluation model. When the matching degree is less than the matching degree threshold, return to perform the operation of determining the target region image from the high-resolution and wide-field image to be detected. Thus, the target region image can be determined from the high-resolution and wide-field image to be detected with a large amount of data, thereby reducing the subsequent data processing volume, improving the stripe detection efficiency, and since the processing complexity of the target region image is relatively low, the stripe detection effect of the high-resolution and wide-field image can be improved.
[0097] Figure 4 It is a schematic structural diagram of a high-resolution and wide-field optical image intelligent stripe detection device proposed in an embodiment of the present disclosure.
[0098] As Figure 4 shown, the high-resolution and wide-field optical image intelligent stripe detection device 40 includes:
[0099] An acquisition module 401, configured to acquire a high-resolution and wide-field image to be detected, where the high-resolution and wide-field image to be detected includes: a plurality of initial image blocks;
[0100] A first determination module 402, configured to determine at least one candidate image block from the plurality of initial image blocks;
[0101] A second determination module 403, configured to determine a first grayscale histogram corresponding to each candidate image block, and according to the first grayscale histogram corresponding to each candidate image block, determine a second grayscale histogram corresponding to the high-resolution and wide-field image to be detected;
[0102] A third determination module 404 is used to determine a target area image from the high-resolution and wide-width image to be detected, and to determine a third grayscale histogram corresponding to the target area image;
[0103] The fourth determination module 405 is used to determine the matching degree between the third grayscale histogram and the second grayscale histogram;
[0104] Processing module 406 is used to input the target area image into the stripe detection model and the quality evaluation model respectively when it is determined that the matching degree is greater than or equal to the matching degree threshold, so as to obtain the stripe detection result output by the stripe detection model and the quality evaluation result output by the quality evaluation model.
[0105] In some embodiments of the present disclosure, the first grayscale histogram includes: a first number of first pixels corresponding to each grayscale level; wherein the second determination module 403 is further configured to:
[0106] Determine a quantitative average of a first quantity corresponding to each gray level in a plurality of first gray histograms;
[0107] A second grayscale histogram corresponding to the high-resolution and wide-width image to be detected is generated according to the average value of the number of first pixel points corresponding to each grayscale level.
[0108] In some embodiments of the present disclosure, the size of the high-resolution and wide-width image to be detected is h*w; wherein the fourth determination module 305 is further used to:
[0109] From the high-resolution and wide-width image to be detected, a local image with a size of x*w located at the center of the image is determined as the target area image, where x <h。
[0110] In some embodiments of the present disclosure, the processing module 406 is further configured to:
[0111] When the matching degree is less than the matching degree threshold, the operation of determining the target area image from the high-resolution and large-width image to be detected is returned, wherein in the operation of determining the target area image from the high-resolution and large-width image to be detected, the image size of the target area image is y*w, x <y。
[0112] In some embodiments of the present disclosure, the model structure of the stripe detection model includes:
[0113] A first standard convolutional module CBL, a first backbone network, a channel-based lightweight recoding module CLR, a second CBL and a second backbone network;
[0114] Among them, the model structure of the quality evaluation model includes:
[0115] A third CBL, a third backbone network, a fourth backbone network, and a fourth CBL.
[0116] In some embodiments of the present disclosure, the stripe detection model is pre-trained based on a first mapping relationship between a sample high-resolution and large-width image and annotated stripe detection results corresponding to the sample high-resolution and large-width image, and the quality evaluation model is pre-trained based on a second mapping relationship between a sample high-resolution and large-width image and annotated signal-to-noise ratio corresponding to the sample high-resolution and large-width image.
[0117] In some embodiments of the present disclosure, the first determining module 402 is further configured to:
[0118] Determining a second number of second pixels in each initial image block;
[0119] When the second number is greater than or equal to the number threshold, the initial image blocks corresponding to the second number are used as candidate image blocks.
[0120] It should be noted that the aforementioned explanation of the intelligent stripe detection method for high-resolution and large-width optical images is also applicable to the intelligent stripe detection device for high-resolution and large-width optical images of the present embodiment, and will not be repeated here.
[0121] In the disclosed embodiment, a high-resolution and large-width image to be detected is obtained, wherein the high-resolution and large-width image to be detected includes: a plurality of initial image blocks, and then at least one candidate image block is determined from the plurality of initial image blocks, and then a first grayscale histogram corresponding to each candidate image block is determined, and then a second grayscale histogram corresponding to the high-resolution and large-width image to be detected is determined according to the first grayscale histogram corresponding to each candidate image block, and then a target area image is determined from the high-resolution and large-width image to be detected, and a third grayscale histogram corresponding to the target area image is determined, and then a matching degree between the third grayscale histogram and the second grayscale histogram is determined, and when it is determined that the matching degree is greater than or equal to a matching degree threshold, the target area image is respectively input into a stripe detection model and a quality evaluation model to obtain a stripe detection result output by the stripe detection model, and a quality evaluation result output by the quality evaluation model, thereby, the target area image can be determined from the high-resolution and large-width image to be detected with a large amount of data, so that the subsequent data processing amount can be reduced, thereby improving the stripe detection efficiency, and because the processing complexity of the target area image is low, the stripe detection effect of the high-resolution and large-width image can be improved.
[0122] Figure 5 A block diagram of an exemplary electronic device suitable for implementing embodiments of the present disclosure is shown. Figure 5 The electronic device 12 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.
[0123] like Figure 5As shown, the electronic device 12 is in the form of a general purpose computing device. The components of the electronic device 12 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, and a bus 18 that connects various system components (including the system memory 28 and the processing unit 16).
[0124] The bus 18 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor or a local bus using any of a variety of bus structures. For example, these architectures include but are not limited to Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MAC) bus, Enhanced ISA bus, Video Electronics Standards Association (VESA) local bus and Peripheral Component Interconnection (PCI) bus.
[0125] The electronic device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the electronic device 12, including volatile and non-volatile media, removable and non-removable media.
[0126] The memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache memory 32. The electronic device 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 34 may be used to read and write non-removable, non-volatile magnetic media ( Figure 5 Not shown, often called a "hard drive").
[0127] although Figure 5Not shown in the figure, a disk drive for reading and writing a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing a removable non-volatile optical disk (e.g., a compact disc read only memory (hereinafter referred to as: CD-ROM), a digital versatile disc read only memory (hereinafter referred to as: DVD-ROM) or other optical media) may be provided. In these cases, each drive may be connected to the bus 18 via one or more data medium interfaces. The memory 28 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the various embodiments of the present disclosure.
[0128] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in the memory 28, such program modules 42 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. The program modules 42 generally perform the functions and / or methods of the embodiments described in the present disclosure.
[0129] The electronic device 12 may also communicate with one or more external devices 14 (e.g., keyboard, pointing device, display 24, etc.), may also communicate with one or more devices that enable the human body to interact with the electronic device 12, and / or communicate with any device that enables the electronic device 12 to communicate with one or more other computing devices (e.g., network card, modem, etc.). Such communication may be performed through an input / output (I / O) interface 22. In addition, the electronic device 12 may also 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) through a network adapter 20. As shown, the network adapter 20 communicates with other modules of the electronic device 12 through a bus 18. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 12, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.
[0130] The processing unit 16 executes various functional applications and data processing by running programs stored in the system memory 28, such as implementing the high-resolution and wide-width optical image intelligent stripe detection method mentioned in the above embodiments.
[0131] In order to implement the above embodiments, the present disclosure also proposes a non-temporary computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the high-resolution and wide-width optical image intelligent stripe detection method proposed in the above embodiments of the present disclosure is implemented.
[0132] In order to implement the above embodiments, the present disclosure further proposes a computer program product. When the instruction processor in the computer program product is executed, the high-resolution and wide-width optical image intelligent stripe detection method proposed in the above embodiments of the present disclosure is executed.
[0133] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. The present disclosure is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The description and examples are to be considered exemplary only, and the true scope and spirit of the present disclosure are indicated by the following claims.
[0134] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.
[0135] It should be noted that, in the description of the present disclosure, the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. In addition, in the description of the present disclosure, unless otherwise specified, the meaning of "plurality" is two or more.
[0136] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present disclosure includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present disclosure belong.
[0137] It should be understood that the various parts of the present disclosure can be implemented in hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.
[0138] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.
[0139] In addition, each functional unit in each embodiment of the present disclosure may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.
[0140] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.
[0141] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present disclosure. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.
[0142] Although the embodiments of the present disclosure have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present disclosure. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present disclosure.
Claims
1. A high-resolution and wide-width optical image intelligent fringe detection method, characterized in that: The method comprises: Acquire a high-resolution and large-width image to be detected, wherein the high-resolution and large-width image to be detected includes: a plurality of initial image blocks; Determine at least one candidate image block from the plurality of initial image blocks; Determine a first grayscale histogram corresponding to each of the candidate image blocks, and determine a second grayscale histogram corresponding to the high-resolution and wide-width image to be detected based on the first grayscale histogram corresponding to each of the candidate image blocks; Determine a target area image from the high-resolution and wide-width image to be detected, and determine a third grayscale histogram corresponding to the target area image; Determining a degree of matching between the third grayscale histogram and the second grayscale histogram; When it is determined that the matching degree is greater than or equal to the matching degree threshold, the target area image is respectively input into the stripe detection model and the quality evaluation model to obtain the stripe detection result output by the stripe detection model and the quality evaluation result output by the quality evaluation model.
2. The method according to claim 1, characterized in that The first grayscale histogram includes: a first number of first pixel points corresponding to each grayscale level; The step of determining, according to the first grayscale histogram corresponding to each of the candidate image blocks, a second grayscale histogram corresponding to the high-resolution and wide-width image to be detected comprises: Determine a quantitative average of a first quantity corresponding to each gray level in a plurality of the first gray histograms; The second grayscale histogram corresponding to the high-resolution and wide-width image to be detected is generated according to the number average of the first pixel points corresponding to each of the grayscale levels.
3. The method according to claim 1, characterized in that The size of the high-resolution and wide-width image to be detected is h*w; Wherein, determining the target area image from the high-resolution and large-width image to be detected includes: From the high-resolution and wide-width image to be detected, a local image with a size of x*w located at the center of the image is determined as the target area image, where x <h。 4. The method according to claim 3, characterized in that The method further comprises: When the matching degree is less than the matching degree threshold, the operation of determining the target area image from the high-resolution and large-width image to be detected is returned, wherein in the operation of determining the target area image from the high-resolution and large-width image to be detected, the image size of the target area image is y*w, x <y。 5. The method according to claim 1, characterized in that The model structure of the fringe detection model includes: A first standard convolutional module CBL, a first backbone network, a channel-based lightweight recoding module CLR, a second CBL and a second backbone network; The model structure of the quality evaluation model includes: A third CBL, a third backbone network, a fourth backbone network, and a fourth CBL.
6. The method according to claim 5, characterized in that The stripe detection model is pre-trained based on a first mapping relationship between a sample high-resolution and large-width image and an annotated stripe detection result corresponding to the sample high-resolution and large-width image, and the quality evaluation model is pre-trained based on a second mapping relationship between the sample high-resolution and large-width image and an annotated signal-to-noise ratio corresponding to the sample high-resolution and large-width image.
7. The method according to any one of claims 1 to 6, characterized in that The step of determining at least one candidate image block from the plurality of initial image blocks comprises: Determine a second number of second pixels in each of the initial image blocks; When the second number is greater than or equal to a number threshold, the initial image blocks corresponding to the second number are used as the candidate image blocks.
8. A high-resolution and wide-width optical image intelligent stripe detection device, characterized in that: The device comprises: An acquisition module is used to acquire a high-resolution and large-width image to be detected, wherein the high-resolution and large-width image to be detected includes: a plurality of initial image blocks; A first determining module, configured to determine at least one candidate image block from a plurality of the initial image blocks; A second determination module is used to determine a first grayscale histogram corresponding to each of the candidate image blocks, and determine a second grayscale histogram corresponding to the high-resolution and wide-width image to be detected based on the first grayscale histogram corresponding to each of the candidate image blocks; A third determination module is used to determine a target area image from the high-resolution and wide-width image to be detected, and determine a third grayscale histogram corresponding to the target area image; A fourth determination module, used to determine a matching degree between the third grayscale histogram and the second grayscale histogram; A processing module is used to input the target area image into a stripe detection model and a quality evaluation model respectively when it is determined that the matching degree is greater than or equal to a matching degree threshold, so as to obtain the stripe detection result output by the stripe detection model and the quality evaluation result output by the quality evaluation model.
9. An electronic device, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 7.
10. A non-transitory computer-readable storage medium storing computer instructions, characterized in that: in, The computer instructions are used to cause the computer to execute the method according to any one of claims 1 to 7.
Citation Information
Patent Citations
Method and system for detecting whether camera is covered or not
CN104240235A
Multi-target interactive image segmentation method and device
CN104899849A
Infrared target detection method based on local sparse representation and contrast
CN106709512A
Target area determination method and system, computer device, and computer storage medium
CN109241862A
A method and a device for extract that center of a laser stripe of a rail profile unde a dynamic driving environment
CN109389639A
