An automatic induction triggering method based on brightness compensation

Abstract

The application relates to an automatic induction triggering method based on brightness compensation, which comprises the following steps: extracting features of a first image and a second image respectively in a plurality of processing domains to obtain a plurality of first features and a plurality of second features; judging whether the second features are modified or not, if the second features are modified, predicting the optimal value of the second features according to the first features; modifying the second features according to the optimal value of the second features; and judging whether a decoding process is triggered or not according to the first features and the second features. The application judges whether the decoding process is triggered or not by using the first features and the modified second features, reduces the influence of image flicker, effectively reduces the number of false touches, and saves the algorithm resource. Furthermore, the optimal value of the second features is predicted according to the first features, the correlation between the first features and the second features is used to improve the accuracy of the second features, the brightness compensation of the second image is realized, and the influence of image flicker is further reduced.

Description

Technical Field

[0001] The invention relates to an automatic induction triggering method based on brightness compensation, and belongs to the field of bar code reading. Background Art

[0002] Barcode readers are widely used across various industries, capable of interpreting barcode images and extracting the string data contained within. Automatic sensing mode is a common operating mode for barcode readers. When a barcode is captured by the device's camera, the device automatically interprets the barcode.

[0003] In the existing technology, the automatic sensing mode is mainly implemented by continuously collecting images through the camera and using algorithms to determine image changes, and then decide whether to trigger the decoding process. The inter-frame difference method is a commonly used algorithm that determines whether there are changes in the image by performing a differential operation on two frames of images. The inter-frame difference method is a global calculation that processes all data in the image. Reasons such as equipment aging and complex working environment lighting can lead to poor image quality, more noise, and increase the number of false positives; on the other hand, due to the large amount of calculation and high complexity of the inter-frame difference method, frequent calls will put the processor under extremely high load, and then generate high heat, resulting in increased equipment failure rate and reduced equipment performance.

[0004] Due to the wide range of application scenarios of barcode reading devices, when the device is in a working environment with complex and extreme light changes, such as strong outdoor sunlight, dim basement parking lots, and indoor environments with multiple scattered light sources, image flickering may occur. That is, the pixels at the same position in two consecutive frame images with the same resolution only have a difference in brightness value, while the objects in the device reading area do not move. In this case, the algorithm may misjudge that a barcode that needs to be parsed appears in the device reading area and trigger the decoding process. When the system is in the decoding process, it cannot parse the barcode that appears later in time, creating the illusion of slow system response and wasting system computing resources.

[0005] Therefore, a barcode reading device induction triggering method with a small number of misjudgments is needed.

[0006] Patent publication number CN104008360B, "An Automatic Sensing Barcode Reading Method and Automatic Sensing Barcode Reader," discloses comparing the sum of the grayscale values ​​of all pixels in the current frame with the sum of the grayscale values ​​of all pixels in the previous frame. If the sum of the grayscale values ​​of all pixels in the current frame differs from the sum of the grayscale values ​​of all pixels in the previous frame, it is determined that the object to be read has entered the reading range. However, this method requires a large amount of computation due to global calculations, and judgments based solely on the sum of the grayscale values ​​of the pixels are prone to misjudgment.

[0007] Patent publication number CN109543487B, "An Automatic Sensing Triggering Method and System Based on Barcode Edge Segmentation," discloses incorporating a barcode edge segmentation algorithm into barcode image preprocessing. This algorithm determines whether the image is the same based on the changes in the white point values ​​of each block after edge segmentation before and after illumination changes, thereby reducing false triggering. However, this method still requires a high level of computational complexity. Summary of the Invention

[0008] To overcome the problems existing in the prior art, the present invention designs an automatic sensing triggering method based on brightness compensation. This method uses a first feature and a modified second feature to determine whether to trigger the decoding process, thereby reducing the impact of image flicker, effectively reducing the number of false triggers, and saving computing resources. Furthermore, the optimal value of the second feature is predicted based on the first feature. The correlation between the first and second features is used to improve the accuracy of the second feature, achieving brightness compensation for the second image, further reducing the impact of image flicker.

[0009] In order to achieve the above object, the present invention adopts the following technical solutions:

[0010] An automatic induction triggering method based on brightness compensation includes the following steps:

[0011] Acquire a first image and a second image;

[0012] Selecting a plurality of processing domains, wherein the processing domain includes at least one pixel point;

[0013] Extracting features from the first image and the second image in the plurality of processing domains to obtain a plurality of first features and a plurality of second features;

[0014] Determine whether to modify the second feature. If the second feature is to be modified, predict an optimal value of the second feature based on the first feature; and modify the second feature based on the optimal value of the second feature.

[0015] According to the first feature and the second feature, it is determined whether to trigger the decoding process.

[0016] Furthermore, the determination of whether the second feature needs to be modified is specifically as follows:

[0017] Perform linear fitting on the plurality of the first features and the second features to obtain a fitted straight line; and determine whether to modify the second feature based on the slope of the fitted straight line.

[0018] Furthermore, the optimal value of the predicted second feature is specifically:

[0019] Perform linear fitting on a plurality of the first features and the second features to obtain a fitting straight line; substitute the first feature into the fitting straight line to obtain an optimal value corresponding to the second feature.

[0020] Furthermore, the determination of whether to trigger the decoding process is specifically as follows:

[0021] Calculate the difference between the first feature and the second feature. If the difference is greater than a preset threshold, enter the decoding process.

[0022] Furthermore, extracting the first feature specifically includes: calculating the cumulative value of the grayscale values ​​of all pixels in the processing domain as the first feature value.

[0023] Furthermore, extracting the first feature specifically includes: eliminating pixels in the processing domain whose grayscale values ​​are greater than a grayscale threshold; and calculating the accumulated value of the grayscale values ​​of the remaining pixels in the processing domain as the first feature value.

[0024] Furthermore, the processing domain is located in the center of the image.

[0025] Furthermore, a plurality of processing domains are provided at the edge and the center of the image.

[0026] Technical Solution 2

[0027] An automatic sensing barcode reader based on brightness compensation, comprising:

[0028] An image acquisition unit, configured to acquire a first image and a second image;

[0029] A trigger judgment unit is provided with a plurality of processing domains, and is used to extract features from the first image and the second image in the plurality of processing domains to obtain a plurality of first features and a plurality of second features; and determine whether to call the barcode reading unit based on the first features and the second features;

[0030] a feature modification unit, the feature modification unit being configured to determine whether to modify the second feature, and if so, predict an optimal value of the second feature based on the first feature; and modify the second feature based on the optimal value of the second feature;

[0031] A barcode reading unit is used to read barcodes.

[0032] Furthermore, the determination of whether the second feature needs to be modified is specifically as follows:

[0033] Perform linear fitting on the plurality of the first features and the second features to obtain a fitted straight line; and determine whether to modify the second feature based on the slope of the fitted straight line.

[0034] Furthermore, the optimal value of the predicted second feature is specifically:

[0035] Perform linear fitting on a plurality of the first features and the second features to obtain a fitting straight line; substitute the first feature into the fitting straight line to obtain an optimal value corresponding to the second feature.

[0036] Furthermore, the determination of whether to trigger the decoding process is specifically as follows:

[0037] Calculate the difference between the first feature and the second feature. If the difference is greater than a preset threshold, enter the decoding process.

[0038] Compared with the prior art, the present invention has the following characteristics and beneficial effects:

[0039] 1. The present invention utilizes the first feature and the modified second feature to determine whether to trigger the decoding process, reducing the impact of image flicker, effectively reducing the number of false triggers, and conserving computing resources. Furthermore, the optimal value of the second feature is predicted based on the first feature, and the correlation between the first and second features is used to improve the accuracy of the second feature, thereby achieving brightness compensation for the second image and further reducing the impact of image flicker.

[0040] 2. The present invention selects several processing domains in the image and performs feature extraction on the processing domains, which reduces the amount of calculation and reduces the interference of invalid data of noise on the algorithm results, which is conducive to increasing the frequency of judging whether to trigger decoding and improving the sensitivity of triggering decoding.

[0041] 3. The present invention determines whether to correct the second feature by the slope of the fitting line, which can effectively determine whether the second feature has a large error, and further improve the accuracy of trigger decoding. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a flow chart of the present invention;

[0043] Figure 2 It is a schematic diagram of the processing domain described in the present invention. DETAILED DESCRIPTION

[0044] The present invention will be described in more detail below with reference to the embodiments.

[0045] In one embodiment, an automatic sensing triggering method based on brightness compensation is characterized by comprising the following steps:

[0046] Acquire a first image and a second image;

[0047] Define several processing domains, each of which includes at least one pixel. In this embodiment, the processing domain position is as follows: Figure 2 The processing domain can be obtained by manual division or by dividing the region of interest through image recognition algorithms.

[0048] In a plurality of processing domains, features are extracted from the first image and the second image to obtain first features and second features. The first features correspond to the second features in a one-to-one manner. In this embodiment, the features are the accumulated grayscale values ​​of all pixels in the processing domain.

[0049] Perform a linear fit on several of the first and second features to obtain a fitted line. If the slope of the fitted line is greater than a slope threshold, it is determined that the second feature needs to be modified. Substitute the several first features into the fitted line to obtain the optimal values ​​of several corresponding second features. Each second feature value is modified to its optimal value. If the slope of the fitted line is less than or equal to the threshold, it is determined that the second feature does not need to be modified and the process proceeds to the next step.

[0050] Calculate the difference between the first feature and its corresponding second feature. If the difference is greater than the difference threshold, enter the decoding process.

[0051] In another embodiment, the optimal value of the second feature is obtained by using data analysis methods such as average analysis, central tendency analysis, and dispersion trend analysis.

[0052] In another embodiment, a region with a width of 20 pixels along the edge of the image is taken as the processing domain, and within the range of the image intermediate resolution of 480*120, 50 regions with a size of 20*20 pixels are uniformly taken as the processing domain.

[0053] When the light is low, the image edge may be lower in brightness than the middle. Calculating only edge features can easily lead to ineffective decoding triggering. This embodiment adds feature extraction to the middle area of ​​the image to improve the accuracy of effective decoding triggering.

[0054] In another embodiment, when calculating the first feature and the second feature, pixels in the processing domain whose grayscale values ​​are greater than the grayscale threshold (the grayscale threshold is 220 in this embodiment) are eliminated, thereby reducing the error impact of extreme values ​​on the features.

[0055] In another embodiment, an automatic sensing triggering method based on brightness compensation includes the following steps:

[0056] The barcode scanning device is fixed and performs multiple image acquisitions at predetermined intervals, from which two adjacent frames of images are selected as the first image and the second image.

[0057] Define n processing domains, such as Figure 2 shown.

[0058] In the n processing domains, features are extracted from the first image and the second image respectively to obtain n first features and n second features. The first features and second features obtained based on the same processing domain correspond to each other.

[0059] For n first features and n second features, take the first feature as the horizontal coordinate and the second coordinate as the vertical coordinate, and get several coordinate points (x1, y1),…(x1, y1)…,(x n ,y n );

[0060] Perform linear fitting on these n coordinate points:

[0061] Assume that the equation of the fitted line is: y=kx+b,

[0062] According to the average value of the first and second characteristics And the slope k, calculate the intercept b and get the equation of the fitted line.

[0063] If the slope k of the fitted line is greater than the threshold Thr1, then modify each second feature: change the n first feature values ​​(x1, ..., x n ) is substituted into the fitted straight line equation to obtain n second characteristic optimal values ​​(y1′,…,y n '); modify each second feature to its optimal value.

[0064] Calculate the absolute value of the difference between the first feature and the second feature |x1-y1′|, ..., |x n -y n '|, if there is at least one absolute value greater than the threshold Thr2, then the decoding process is entered to decode the image obtained after the first image.

[0065] Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

Claims

1. An automatic induction triggering method based on brightness compensation, characterized in that: The following steps are involved: Acquire a first image and a second image; Selecting a plurality of processing domains, wherein the processing domain includes at least one pixel point; Extracting features from the first image and the second image in the plurality of processing domains to obtain a plurality of first features and a plurality of second features; The first feature is extracted as follows: Eliminate pixels whose grayscale values ​​in the processing domain are greater than the grayscale threshold; calculate the accumulated value of the grayscale values ​​of the remaining pixels in the processing domain as the first eigenvalue; Performing linear fitting on a plurality of the first features and the second features to obtain a fitting straight line; Determine whether to modify the second feature based on the slope of the fitted line. If the second feature is to be modified, substitute the first feature into the fitted line to obtain the optimal value of the second feature. Modify the second feature based on the optimal value of the second feature. Calculate the difference between the first feature and the second feature. If the difference is greater than a preset threshold, enter the decoding process.

2. An automatic sensing barcode reader based on brightness compensation, characterized in that: include: An image acquisition unit, configured to acquire a first image and a second image; A trigger judgment unit is provided with a plurality of processing domains, and is used to extract features from the first image and the second image in the plurality of processing domains to obtain a plurality of first features and a plurality of second features; and determine whether to call the barcode reading unit based on the first features and the second features; The first feature is extracted as follows: Eliminate pixels whose grayscale values ​​in the processing domain are greater than the grayscale threshold; calculate the accumulated value of the grayscale values ​​of the remaining pixels in the processing domain as the first eigenvalue; Performing linear fitting on a plurality of the first features and the second features to obtain a fitting straight line; a feature modification unit, wherein the feature modification unit determines whether to modify the second feature based on the slope of the fitted line, and if the second feature is to be modified, the first feature is substituted into the fitted line to obtain an optimal value of the second feature; and the second feature is modified based on the optimal value of the second feature; The barcode reading unit is used to calculate the difference between the first feature and the second feature, and read the barcode if the difference is greater than a preset threshold.

Images