High-speed photographic apparatus image data processing method and system

Abstract

The invention discloses a high-speed photographic apparatus image data processing method and system, and the method comprises the steps: obtaining collected high-speed photographic apparatus image information, and carrying out the preprocessing of the high-speed photographic apparatus image information; combining the preprocessed image information of the high-speed photographic apparatus with a pre-constructed messy code recognition model to obtain a recognition result for converting the image into a messy code document; and performing corresponding image processing and / or reacquisition processing on the image information of the high-speed photographic apparatus according to the identification result. According to the method, the image information which cannot be successfully converted into the document can be efficiently and comprehensively identified and processed, and the image processing efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of high-speed scanners, and in particular to a method and system for processing image data of high-speed scanners. Background Art

[0002] Gaopaiyi, also known as the speed photo instrument or the lesson preparation king, is a newly developed ultra-portable and low-carbon office supply. It is also an innovative product in the office field. It has a foldable and ultra-convenient design, can complete high-speed scanning in one second, and has OCR text recognition function, which can convert scanned images into editable documents.

[0003] When the high-speed scanner converts the scanned image information into editable document information, garbled characters may appear in the converted document information. The reason for this is that the captured image is not clear or the shooting environment is too dark, resulting in a dark image. Therefore, after the high-speed scanner acquires the image, the image needs to be processed.

[0004] The existing processing method mainly processes the acquired images based on experience, which is inefficient and may also result in incomplete processing. Summary of the invention

[0005] In order to solve the above technical problems, an embodiment of the present invention provides a method for processing image data of a high-speed scanner, comprising:

[0006] Acquire the collected high-speed scanner image information and preprocess the high-speed scanner image information; combine the preprocessed high-speed scanner image information with a pre-built garbled code recognition model to obtain a recognition result that causes the image to be converted into a garbled code document; perform corresponding image processing and / or re-collection processing on the high-speed scanner image information according to the recognition result.

[0007] Preferably, the garbled code recognition model is constructed according to the following steps: obtain a first historical gps image that is successfully converted into a document and a second historical gps image that is not converted into a document; based on the first historical gps image and the second historical gps image, divide them according to a preset ratio to obtain a training set and a verification set; based on the training set, train using a neural network model to obtain an initial recognition model; based on the verification set, adjust the feature information threshold in the initial recognition model, and construct a garbled code recognition model based on the initial recognition model after adjusting the feature information threshold.

[0008] Preferably, in the step of training using a neural network model based on the training set to obtain an initial recognition model, the step includes: based on the training set, using the neural network model to extract historical camera image information in the training set to obtain first feature information and second feature information; based on the first feature information and the second feature information, obtaining a feature information threshold that is successfully converted to the document; based on the feature information threshold, in combination with the neural network model, constructing an initial recognition model.

[0009] Preferably, the step of adjusting the feature information threshold in the initial recognition model based on the verification set includes: inputting the images in the verification set into the initial recognition model to obtain recognition results corresponding to the images in the verification set; when the accuracy of the recognition results corresponding to the images in the verification set is less than or equal to a preset value, correcting the feature information threshold in the initial model.

[0010] Preferably, the feature information includes but is not limited to: image pixel features, image edge blur, image brightness features and image contrast features; wherein, in combining the preprocessed high-speed scanner image with a pre-built garbled code recognition model, the step of obtaining a recognition result that causes the image to be converted into a garbled code document includes: extracting information from the preprocessed high-speed scanner image to obtain image pixel features, image edge blur, image brightness features and image contrast features corresponding to the high-speed scanner image; based on the image pixel features, image edge blur, image brightness features and image contrast features and the feature information threshold in the garbled code recognition model, obtaining a recognition result that causes the image to be converted into a garbled code document.

[0011] Preferably, the feature information threshold includes: an image feature threshold, an image edge blur threshold and an exposure threshold; wherein, in the step of obtaining the recognition result causing the image to be converted into a garbled document based on the image pixel feature, image edge blur, image brightness feature and image contrast feature and the feature information threshold in the garbled code recognition model, it includes: when the image pixel feature is less than or equal to the image feature threshold, the recognition result causing the image to be converted into a garbled code document is that the high-speed image sensor image is too dark; when the image edge blur is less than or equal to the image edge blur threshold, the recognition result causing the image to be converted into a garbled code document is that the high-speed image sensor image is a blurred image; when the value after the image brightness feature and the image contrast feature are coupled is greater than the exposure threshold, the recognition result causing the image to be converted into a garbled code document is that the high-speed image sensor image is overexposed.

[0012] Preferably, the step of performing corresponding image processing and / or re-capture processing on the high-definition camera image according to the recognition result includes: according to the recognition result, using an image correction algorithm to correct the high-definition camera image to obtain an image that successfully converts the document information; when the corrected high-definition camera image fails to be successfully converted into document information, re-capture processing is performed.

[0013] Preferably, when the corrected high-speed scanner image fails to be successfully converted into document information, the step of re-capture processing includes: based on the recognition result, correcting the shooting environment of the high-speed scanner until the feature information of the preview image obtained based on the corrected high-speed scanner environment information meets the preset conditions.

[0014] On the other hand, an embodiment of the present invention further provides a high-speed image data processing system, including:

[0015] An image processing module is used to obtain the collected high-speed scanner image information and pre-process the high-speed scanner image information; a recognition result acquisition module is used to combine the pre-processed high-speed scanner image information with a pre-built garbled code recognition model to obtain a recognition result that causes the image to be converted into a garbled code document; a result processing module is used to perform corresponding image processing and / or re-collection processing on the high-speed scanner image information according to the recognition result.

[0016] Preferably, the garbled code recognition model is constructed according to the following steps: obtain a first historical gps image that is successfully converted into a document and a second historical gps image that is not converted into a document; based on the first historical gps image and the second historical gps image, divide them according to a preset ratio to obtain a training set and a verification set; based on the training set, train using a neural network model to obtain an initial recognition model; based on the verification set, adjust the feature information threshold in the initial recognition model, and construct a garbled code recognition model based on the initial recognition model after adjusting the feature information threshold.

[0017] Compared with the prior art, one or more embodiments of the above scheme may have the following advantages or beneficial effects:

[0018] The present invention discloses a method and system for processing image data of a high-speed scanner, including: obtaining collected high-speed scanner image information, and preprocessing the high-speed scanner image information; combining the preprocessed high-speed scanner image information with a pre-built garbled code recognition model to obtain a recognition result that causes the image to be converted into a garbled code document; and performing corresponding image processing and / or re-collection processing on the high-speed scanner image information according to the recognition result. The present invention can efficiently and comprehensively recognize and process image information that has not been successfully converted into a document, thereby improving the efficiency of image processing.

[0019] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0021] Figure 1 Schematic diagram of the overall steps of the method for processing high-speed camera image data according to an embodiment of the present application.

[0022] Figure 2 Schematic diagram of the overall structure of the method for processing high-speed camera image data according to an embodiment of the present application. DETAILED DESCRIPTION

[0023] The following will describe the implementation methods of the present invention in detail with reference to the accompanying drawings and embodiments, so that the implementation process of how the present invention applies technical means to solve technical problems and achieve technical effects can be fully understood and implemented accordingly. It should be noted that as long as there is no conflict, the various embodiments of the present invention and the various features in the embodiments can be combined with each other, and the technical solutions formed are all within the protection scope of the present invention.

[0024] In addition, the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions. Also, although a logical sequence is shown in the flowchart, in some cases, the steps shown or described can be performed in a sequence different from that here.

[0025] The terms used herein are only for describing specific embodiments and are not intended to limit exemplary embodiments. Unless the context clearly indicates otherwise, the singular forms "one", "one" and "item" used herein are also intended to include plural numbers. It should also be understood that the terms "include" and / or "comprise" used herein specify the existence of stated features, integers, steps, operations, units and / or components, without excluding the existence or addition of one or more other features, integers, steps, operations, units, components and / or combinations thereof.

[0026] Gaopaiyi, also known as the speed photo instrument or the lesson preparation king, is a newly developed ultra-portable and low-carbon office supply. It is also an innovative product in the office field. It has a foldable and ultra-convenient design, can complete high-speed scanning in one second, and has OCR text recognition function, which can convert scanned images into editable documents.

[0027] When the high-speed scanner converts the scanned image information into editable document information, garbled characters may appear in the converted document information. The reason for this is that the captured image is not clear or the shooting environment is too dark, resulting in a dark image. Therefore, after the high-speed scanner acquires the image, the image needs to be processed.

[0028] The existing processing method mainly processes the acquired images based on experience, which is inefficient and may also result in incomplete processing.

[0029] Embodiment 1

[0030] Figure 1 This is a schematic diagram of the overall steps of the method for processing image data of a high-speed camera according to an embodiment of the present application. Figure 1 The specific process of the method for processing high-speed image data according to the embodiment of the present invention is described.

[0031] like Figure 1 As shown, the high-speed camera image data processing method includes:

[0032] Step S110, acquiring the collected high-speed scanner image information, and preprocessing the high-speed scanner image information.

[0033] Step S120, combining the pre-processed high-speed scanner image information with the pre-built garbled code recognition model to obtain a recognition result that causes the image to be converted into a garbled code document.

[0034] Step S130: performing corresponding image processing and / or re-capture processing on the high-speed camera image information according to the recognition result.

[0035] In this embodiment, the pre-processing method of the high-speed scanner image includes but is not limited to: denoising, normalization and histogram equalization processing to improve the image quality and recognition accuracy.

[0036] In this embodiment, the recognition results that cause the image to be converted into a garbled document include: the high-definition camera image is too dark, the high-definition camera image is a blurred image, and the high-definition camera image is overexposed.

[0037] In this embodiment, the image processing method includes but is not limited to: a brightness adjustment algorithm, a contrast adjustment algorithm, a deblurring algorithm, a sharpening process and a color correction process.

[0038] In this embodiment, the garbled code recognition model can be obtained by training the historical high-speed camera images using a neural network model. Specifically, the garbled code recognition model is constructed according to the following steps A1-A4:

[0039] Step A1, obtaining a first historical high-speed camera image that is successfully converted into a document and a second historical high-speed camera image that is not converted into a document;

[0040] Step A2, based on the first historical high-speed camera image and the second historical high-speed camera image, dividing them according to a preset ratio to obtain a training set and a verification set;

[0041] Step A3, based on the training set, using the neural network model to train and obtain an initial recognition model;

[0042] Step A4: Based on the validation set, the feature information threshold in the initial recognition model is adjusted, and a garbled code recognition model is constructed based on the initial recognition model after the feature information threshold is adjusted.

[0043] In step A2, the specific value of the preset ratio is not limited and can be reasonably selected according to actual application requirements. For example, the preset ratio of the training set to the validation set can be 7:3, 6:4, or 8:2.

[0044] Among them, in step A3, the information of historical high-speed scanner images in the training set is extracted, and based on the feature information, the feature information threshold corresponding to the document for successful conversion of different features is obtained, and then the initial recognition model is constructed according to the feature information threshold.

[0045] Specifically, step A3 includes the following sub-steps A31-A33:

[0046] Sub-step A31, based on the training set, using the neural network model to extract the historical high-speed camera image information in the training set to obtain the first feature information and the second feature information;

[0047] Sub-step A32, obtaining a threshold value of feature information corresponding to a document successfully converted based on the first feature information and the second feature information;

[0048] Sub-step A33, constructing an initial recognition model based on the feature information threshold and in combination with the neural network model.

[0049] In step A4, the feature information includes but is not limited to: image pixel feature, image edge blur, image brightness feature and image contrast feature. The feature information threshold includes: image feature threshold, image edge blur threshold and exposure threshold.

[0050] Optionally, in order to make the obtained feature information threshold more accurate, a verification set is used to verify and calibrate the threshold. Specifically, step A4 includes the following sub-steps A41-A42:

[0051] Sub-step A41, inputting the images in the verification set into the initial recognition model to obtain the recognition results corresponding to the images in the verification set;

[0052] Sub-step A42, when the accuracy of the recognition result corresponding to the image in the verification set is less than or equal to the preset value, correct the feature information threshold in the initial model.

[0053] There is no limit to the specific value of the preset value, and it can be reasonably selected according to actual application requirements.

[0054] In this embodiment, in the step of obtaining the recognition result of converting the image into a garbled document, the preprocessed high-speed image information is combined with the garbled recognition model to obtain image feature information, and the image feature information is compared with the feature information threshold, and the recognition result is obtained based on the comparison result. Specifically, step S120 includes the following sub-steps B1-B2:

[0055] Sub-step B1, extracting information from the pre-processed high-definition camera image to obtain image pixel features, image edge blur, image brightness features, and image contrast features corresponding to the high-definition camera image;

[0056] Sub-step B2, based on the image pixel features, image edge blur, image brightness features, image contrast features and the feature information threshold in the garbled code recognition model, obtain the recognition result that causes the image to be converted into a garbled code document.

[0057] Optionally, sub-step B2 includes the following steps B21-B23:

[0058] Step B21, when the image pixel feature is less than or equal to the image feature threshold, the recognition result that causes the image to be converted into a garbled document is that the high-speed scanner image is too dark;

[0059] Step B22, when the image edge blur is less than or equal to the image edge blur threshold, the recognition result that causes the image to be converted into a garbled document is that the high-speed scanner image is a blurred image;

[0060] Step B23, when the value after the image brightness feature and the image contrast feature are coupled is greater than the exposure threshold, the recognition result that causes the image to be converted into a garbled document is that the high-speed camera image is overexposed.

[0061] In this embodiment, after obtaining the recognition result, the image correction algorithm is used to correct the high-speed scanner image. When the corrected image fails to be successfully converted into document information, resampling processing is performed. Specifically, step S130 includes the following sub-steps C1-C2:

[0062] Sub-step C1, based on the recognition result, the high-resolution image is corrected using an image correction algorithm to obtain an image that successfully converts the document information.

[0063] Sub-step C2: when the corrected high-definition camera image cannot be successfully converted into document information, re-capture processing is performed.

[0064] Among them, in sub-step C1, when the high-definition camera image is too dark, the high-definition camera image can be adjusted by using a brightness adjustment algorithm, sharpening processing and color correction processing to improve the clarity of the high-definition camera image; when the high-definition camera image is a blurred image, the high-definition camera image can be adjusted by using a deblurring algorithm to restore the details and clarity of the high-definition camera image; when the high-definition camera image is overexposed, a re-capture process can be used.

[0065] In the process of re-collection, the environment information of the high-speed scanner must be adjusted so that the acquired high-speed scanner image can be successfully converted into document information. Specifically, sub-step C2 includes:

[0066] Based on the recognition result, the shooting environment of the high-speed camera is corrected until the feature information of the preview image obtained based on the corrected high-speed camera environment information meets the preset conditions.

[0067] Among them, the preset conditions include: the image pixel feature is greater than the image feature threshold, the image edge blur is greater than the image edge blur threshold, and the value after the image brightness feature and the image contrast feature are coupled is less than or equal to the exposure threshold.

[0068] Embodiment 2

[0069] Based on the high-speed scanner image data processing method provided in the first embodiment, the embodiment of the present invention further provides a high-speed scanner image data processing system.

[0070] Figure 2 FIG. 1 is a schematic diagram of the overall structure of the high-speed camera image data processing system according to an embodiment of the present application. Figure 2 As shown, the high-speed camera image data processing system described in the embodiment of the present invention includes:

[0071] An image processing module is used to obtain the collected high-speed camera image information and pre-process the high-speed camera image information;

[0072] The recognition result acquisition module is used to combine the pre-processed high-speed scanner image information with the pre-built garbled code recognition model to obtain the recognition result that causes the image to be converted into a garbled code document;

[0073] The result processing module is used to perform corresponding image processing and / or re-collection processing on the high-speed scanner image information according to the recognition result.

[0074] In one embodiment, the garbled code recognition model is constructed according to the following steps:

[0075] Acquire a first historical camera image that is successfully converted into a document and a second historical camera image that is not converted into a document;

[0076] Based on the first historical high-speed camera image and the second historical high-speed camera image, dividing them according to a preset ratio to obtain a training set and a verification set;

[0077] Based on the training set, the neural network model is used for training to obtain the initial recognition model;

[0078] Based on the validation set, the feature information threshold in the initial recognition model is adjusted, and a garbled code recognition model is constructed based on the initial recognition model after adjusting the feature information threshold.

[0079] The present invention discloses a method and system for processing image data of a high-speed scanner, including: obtaining collected high-speed scanner image information, and preprocessing the high-speed scanner image information; combining the preprocessed high-speed scanner image information with a pre-built garbled code recognition model to obtain a recognition result that causes the image to be converted into a garbled code document; and performing corresponding image processing and / or re-collection processing on the high-speed scanner image information according to the recognition result. The present invention can efficiently and comprehensively recognize and process image information that has not been successfully converted into a document, thereby improving the efficiency of image processing.

[0080] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by any person familiar with the technology within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0081] In the description of the present invention, unless otherwise specified, "plurality" means two or more than two; the orientations or positional relationships indicated by the terms "upper", "lower", "left", "right", "inner", "outer", "front end", "rear end", "head", "tail", etc. are based on the orientations or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operate in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", "third", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0082] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "connected" and "connection" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0083] It should be understood that the embodiments disclosed in the present invention are not limited to the specific structures, processing steps or materials disclosed herein, but should be extended to equivalent substitutions of these features understood by ordinary technicians in the relevant field. It should also be understood that the terms used herein are only used for the purpose of describing specific embodiments and are not meant to be limiting.

[0084] The "one embodiment" or "embodiment" mentioned in the specification means that the specific features, structures or characteristics described in conjunction with the embodiment are included in at least one embodiment of the present invention. Therefore, the phrases "one embodiment" or "embodiment" appearing in various places throughout the specification do not necessarily refer to the same embodiment.

[0085] Although the embodiments disclosed in the present invention are as above, the contents described are only embodiments adopted for facilitating the understanding of the present invention and are not intended to limit the present invention. Any technician in the technical field to which the present invention belongs can make any modifications and changes in the form and details of the implementation without departing from the spirit and scope disclosed in the present invention, but the patent protection scope of the present invention shall still be subject to the scope defined in the attached claims.

Claims

1. A method for processing high-speed image data, characterized in that: The method comprises: Acquire the collected high-speed camera image information and pre-process the high-speed camera image information; The pre-processed high-speed scanner image information is combined with the pre-built garbled code recognition model to obtain the recognition result that causes the image to be converted into a garbled code document; The high-definition camera image information is subjected to corresponding image processing and / or re-collection processing according to the recognition result.

2. The method according to claim 1, characterized in that: The garbled code recognition model is constructed according to the following steps: Acquire a first historical camera image that is successfully converted into a document and a second historical camera image that is not converted into a document; Based on the first historical high-speed camera image and the second historical high-speed camera image, dividing them according to a preset ratio to obtain a training set and a verification set; Based on the training set, a neural network model is used for training to obtain an initial recognition model; Based on the verification set, the feature information threshold in the initial recognition model is adjusted, and a garbled code recognition model is constructed based on the initial recognition model after the feature information threshold is adjusted.

3. The method according to claim 2, characterized in that The step of training the neural network model based on the training set to obtain an initial recognition model includes: Based on the training set, extracting historical high-speed camera image information in the training set using the neural network model to obtain first feature information and second feature information; Based on the first feature information and the second feature information, a feature information threshold corresponding to a document successfully converted is obtained; Based on the feature information threshold and in combination with a neural network model, an initial recognition model is constructed.

4. The method according to any one of claims 2 to 3, characterized in that: The step of adjusting the feature information threshold in the initial recognition model based on the verification set includes: The images in the verification set are fed into the initial recognition model to obtain recognition results corresponding to the images in the verification set; When the accuracy of the recognition result corresponding to the image in the verification set is less than or equal to a preset value, the feature information threshold in the initial model is corrected.

5. The method according to claim 4, characterized in that The feature information includes but is not limited to: image pixel features, image edge blur, image brightness features and image contrast features; wherein, the step of combining the pre-processed high-speed scanner image with the pre-built garbled code recognition model to obtain the recognition result of converting the image into a garbled code document includes: Extracting information from the preprocessed high-definition camera image to obtain image pixel features, image edge blur, image brightness features, and image contrast features corresponding to the high-definition camera image; Based on the image pixel features, image edge blur, image brightness features, image contrast features and the feature information threshold in the garbled code recognition model, a recognition result that causes the image to be converted into a garbled code document is obtained.

6. The method according to claim 5, characterized in that The feature information threshold includes: an image feature threshold, an image edge blur threshold and an exposure threshold; wherein, in the step of obtaining a recognition result that causes the image to be converted into a garbled document based on the image pixel feature, image edge blur, image brightness feature and image contrast feature and the feature information threshold in the garbled code recognition model, the step includes: When the image pixel feature is less than or equal to the image feature threshold, the recognition result that causes the image to be converted into a garbled document is that the high-speed camera image is too dark; When the image edge blur is less than or equal to the image edge blur threshold, the recognition result that causes the image to be converted into a garbled document is that the high-speed camera image is a blurred image; When the value of the image brightness feature and the image contrast feature coupled is greater than the exposure threshold, the recognition result that causes the image to be converted into a garbled document is that the high-speed camera image is overexposed.

7. The method according to claim 6, characterized in that The step of performing corresponding image processing and / or re-collection processing on the high-speed camera image according to the recognition result includes: According to the recognition result, the high-speed scanner image is corrected by using an image correction algorithm to obtain an image that successfully converts the document information; When the corrected high-speed scanner image cannot be successfully converted into document information, re-capture processing is performed.

8. The method according to claim 7, characterized in that When the corrected high-speed scanner image fails to be successfully converted into document information, the steps of re-capturing the image include: Based on the recognition result, the shooting environment of the high-speed scanner is corrected until the feature information of the preview image obtained based on the corrected high-speed scanner environment information meets the preset conditions.

9. A high-speed camera image data processing system, characterized in that: The system comprises: An image processing module is used to obtain the collected high-speed camera image information and pre-process the high-speed camera image information; The recognition result acquisition module is used to combine the pre-processed high-speed scanner image information with the pre-built garbled code recognition model to obtain the recognition result that causes the image to be converted into a garbled code document; The result processing module is used to perform corresponding image processing and / or re-collection processing on the high-speed scanner image information according to the recognition result.

10. The system according to claim 9, characterized in that The garbled code recognition model is constructed according to the following steps: Acquire a first historical camera image that is successfully converted into a document and a second historical camera image that is not converted into a document; Based on the first historical high-speed camera image and the second historical high-speed camera image, dividing them according to a preset ratio to obtain a training set and a verification set; Based on the training set, a neural network model is used for training to obtain an initial recognition model; Based on the verification set, the feature information threshold in the initial recognition model is adjusted, and a garbled code recognition model is constructed based on the initial recognition model after the feature information threshold is adjusted.

Images