Material coil allowance detection method, device and equipment, storage medium and series welding machine

Abstract

The invention provides a material coil allowance detection method and device, equipment, a storage medium and a series welding machine. The method comprises the steps that a material coil image during discharging is collected; intercepting a feature image containing a material roll area from the material roll image; extracting features of the feature image and obtaining feature values; the state of the current material roll is judged based on the characteristic value, wherein the material roll state at least comprises material roll full disc and material roll exhaustion; when the material roll is in the full-disc state, the step of collecting the material roll image during discharging continues to be executed; when the material roll is in the material roll exhaustion state, first early warning information used for prompting replacement of the material roll is output; the method comprises the following steps of: after acquiring a material roll image when a solder strip material roll is discharged, extracting features in a feature image containing a material roll area in the material roll image to obtain feature data, judging whether the current material roll state is full of the material roll or the material roll is used up according to a feature value, and outputting first early warning information for prompting to replace the material roll when the material roll is used up. And the early warning prompt can be carried out when the welding strip is used up without manual adjustment.

Description

Technical Field

[0001] The present application relates to the technical field of photovoltaic module manufacturing, and in particular to a method, device, equipment, storage medium and string welding machine for detecting the remaining amount of a material roll. Background Art

[0002] String welding is an important process in the production of photovoltaic cell modules. The string welding machine welds the welding ribbon to the photovoltaic cell to form a cell string. The welding ribbon wrapped on the roll is pulled, cut and stacked with the photovoltaic cell. The stacked photovoltaic cell and the welding ribbon are connected in series to form a cell string under the action of the stringing device. When the welding ribbon on the roll of the string welding machine is exhausted, it is necessary to replace it with a new roll in time.

[0003] At present, the detection of welding strips is to install a proximity switch at the position of each material coil on the string welding machine to detect the exhaustion of welding strips. The proximity switch can detect the metal welding wire on the material coil. When the welding wire is exhausted, the proximity switch cannot detect the welding wire and will alarm. However, this method has at least the following defects: each material coil needs to be equipped with a proximity switch. Now each string welding machine needs to use up to dozens of proximity switches, which is costly; the detection range is small and cannot achieve full range detection. When the welding strip is almost exhausted, manual adjustment is required; the detection range is single and can only detect the exhaustion of metal materials. Summary of the invention

[0004] In order to solve the problems of the prior art, the present application provides a material coil remainder detection method, device, equipment, storage medium and string welding machine, which can solve the current problems of difficult deployment, high cost and low detection accuracy of solder strip material coil exhaustion detection.

[0005] In a first aspect, the present application provides a method for detecting the remaining amount of a roll, the method comprising:

[0006] Collect the image of the material roll when unwinding;

[0007] Extracting a feature image including a roll area from the roll image;

[0008] Extract the features of the feature image and obtain the feature value;

[0009] Based on the characteristic value, the state of the current material roll is judged, and the material roll state at least includes the material roll being full or the material roll being exhausted;

[0010] When the material roll is in a full reel state, the step of collecting the image of the material roll during unwinding is continued;

[0011] When the material roll is in a state of being exhausted, a first warning message for prompting replacement of the material roll is output.

[0012] The method for detecting the remaining amount of a material coil provided by the present application, after obtaining the material coil image when the solder strip material coil is unwinding, extracts features from a feature image containing the material coil area in the material coil image to obtain feature data, and judges whether the current material coil status is a full coil or an exhausted coil according to the feature value, and outputs a first warning message prompting replacement of the material coil when the material coil is in an exhausted coil status, and can issue a warning prompt when the solder strip on the material coil is about to run out without manual adjustment.

[0013] In some embodiments, the roll status further includes a roll-out status, and the method further includes:

[0014] When the material roll is in a state of being about to run out, a second warning message indicating that the material roll is about to run out is output, and the step of collecting the image of the material roll during unwinding is continued; or,

[0015] When the material roll is in a state where the material roll is about to run out, a second warning message indicating that the material roll is about to run out is output.

[0016] When feature data is obtained based on features in a feature image that includes a material roll area and it is determined based on the feature value that the current material roll is about to run out, a second warning message indicating that the material roll is about to run out is output to issue a warning, and the step of continuing to collect images of the material roll during unwinding is performed to perform continuous real-time detection of the current material roll, or a direct warning is issued and then real-time detection of the replaced current material roll is performed after replacement.

[0017] In some embodiments, extracting features of the feature image and obtaining feature values ​​includes:

[0018] Perform noise reduction and grayscale processing on the feature image;

[0019] The features of the feature image after noise reduction and grayscale processing are extracted, and the feature values ​​are calculated based on the extracted features.

[0020] Before extracting the features of the feature image and calculating the feature values, the feature image is subjected to noise reduction and grayscale processing to improve the quality of the feature image and reduce interference with feature extraction.

[0021] In some embodiments, extracting features of the feature image after noise reduction and grayscale processing, and calculating feature values ​​according to the extracted features, includes:

[0022] Acquire the grayscale value of each pixel in the first area unit from the feature image, and obtain the first feature value according to the grayscale value of each pixel in the first area unit;

[0023] The status of the current roll is determined based on the characteristic values, including:

[0024] If the first characteristic value belongs to the first value range, it is determined that the material roll is in a full reel state;

[0025] If the first characteristic value belongs to the second value range, it is determined that the material roll is in a material roll exhaustion state.

[0026] The grayscale value of each pixel in the first area unit, i.e., at one end of the material roll, is obtained from the feature image, and the first feature value is calculated based on the grayscale value of the pixel. The state of the current material roll is judged based on the first feature value at the end of the material roll and a preset value range, so that the state of the material roll can be detected quickly and accurately.

[0027] In some embodiments, extracting features of the feature image after noise reduction and grayscale processing, and calculating feature values ​​according to the extracted features, includes:

[0028] Acquire the grayscale value of each pixel in the first area unit from the feature image, and obtain the first feature value according to the grayscale value of each pixel in the first area unit;

[0029] Acquire the gray value of each pixel point in the second area unit from the characteristic image, and obtain the second characteristic value according to the gray value of each pixel point in the second area unit, wherein the first area unit and the second area unit are arranged along the axial direction of the roll;

[0030] The status of the current roll is determined based on the characteristic values, including:

[0031] If both the first characteristic value and the second characteristic value belong to the first value range, it is determined that the material roll is in a full reel state;

[0032] If the first characteristic value and the second characteristic value both belong to the second value range, it is determined that the material roll is in a material roll exhaustion state, and the ranges of the first value range and the second value range are different.

[0033] By calculating the characteristic values ​​of the areas at both ends of the roll according to the gray values ​​at both ends of the roll, it is determined whether there are welding lines at both ends of the roll according to the characteristic values ​​of the areas at both ends of the roll, thereby more accurately determining the state of the current roll.

[0034] In some embodiments, the roll status also includes a roll-out status.

[0035] The second area unit is located between the first edge of the roll and the extraction area of ​​the first area unit, and the state of the current roll is judged based on the characteristic value, and further includes:

[0036] When the first characteristic value is within the first value range and the second characteristic value is within the second value range, it is determined that the roll is exhausted in a direction away from the first edge of the roll;

[0037] When the first characteristic value is within the second value range and the second characteristic value is within the first value range, it is determined that the material roll is exhausted in a direction toward the first edge of the material roll;

[0038] or,

[0039] The extraction area of ​​the first area unit is located between the first edge of the roll and the extraction area of ​​the second area unit, and the state of the current roll is judged based on the characteristic value, further comprising:

[0040] When the first characteristic value is within the first value range and the second characteristic value is within the second value range, it is determined that the roll is exhausted in a direction toward the first edge of the roll;

[0041] When the first characteristic value is within the second value range and the second characteristic value is within the first value range, it is determined that the material roll is exhausted in a direction away from the first edge of the material roll.

[0042] According to the value range of the point features in the frame in the image and the point features outside the frame in the image, the material usage of the solder strip inside and outside the frame in the image can be determined, so that the direction in which the solder strip is used up can be accurately determined.

[0043] In some embodiments, after determining that the web is exhausted in a direction toward the first edge of the web, the method further comprises:

[0044] If the characteristic image is close to the first edge of the material roll, it is determined that the material roll is in a state of being exhausted; if the characteristic image is far away from the first edge of the material roll, it is determined that the material roll is in a state of being about to be exhausted;

[0045] After determining that the roll is exhausted in a direction away from the first edge of the roll, the method further includes:

[0046] If the characteristic image is close to the first edge of the material roll, it is determined that the material roll is in a state of being about to run out; if the characteristic image is far away from the first edge of the material roll, it is determined that the material roll is in a state of being run out.

[0047] After determining the direction of the current roll being used up, the specific state of the roll is determined based on whether the characteristic image is close to the first edge of the roll or far from the first edge of the roll, so that the result of determining the state of the roll is more accurate.

[0048] In some embodiments, the method for detecting the remaining amount of the roll further includes:

[0049] Acquire the grayscale value of each pixel point in the third area unit outside the characteristic image along the axial direction of the material roll from the material roll image, and obtain the third characteristic value according to the grayscale value of each pixel point in the third area unit;

[0050] The third area unit is located between the first edge of the roll and the characteristic image. When the roll runs out in the direction away from the first edge of the roll, the third characteristic value is in the first value range, and the first characteristic value and the second characteristic value are in the second value range, an error is reported; when the roll runs out in the direction toward the first edge of the roll, the third characteristic value is in the second value range, and the first characteristic value and the second characteristic value are in the first value range, an error is reported;

[0051] or,

[0052] The characteristic image is located between the first edge of the material roll and the third area unit. When the material roll is used up in the direction toward the first edge of the material roll, the third characteristic value is in the first value range, and the first characteristic value and the second characteristic value are in the second value range, an error is reported; when the material roll is used up in the direction away from the first edge of the material roll, the third characteristic value is in the second value range, and the first characteristic value and the second characteristic value are in the first value range, an error is reported.

[0053] When it is judged that the solder ribbon inside the image frame is used up but the solder ribbon outside the frame is not used up based on the value range of the point features inside the image frame and the point features outside the image frame, an error reporting process is performed to achieve correction of the material roll remainder detection and avoid erroneous detection affecting production efficiency.

[0054] In some embodiments, after extracting a feature image including the roll area from the roll image, the method further comprises:

[0055] Use edge detection to detect the edges of objects in the feature image, segment the feature image to obtain a segmented image, and extract the features of the segmented image;

[0056] Inputting the features of the segmented image into a predetermined matching model, wherein a reference image including a roll area is pre-stored in the predetermined matching model;

[0057] If the predetermined matching model determines that the segmented image matches the reference image successfully, then continue to perform the step of extracting features of the feature image and obtaining feature values;

[0058] If the predetermined matching model determines that the segmented image and the reference image are not matched successfully, an error message is given.

[0059] Combined with the preset coordinates and image edge segmentation information, a static image of the preset identification points on the pad is segmented and obtained; a predetermined matching model is used to determine whether the segmented static image can be successfully matched with the reference image, thereby judging whether the segmented static image can meet the detection requirements, avoiding invalid detection of images that do not meet the requirements and reducing detection efficiency.

[0060] In a second aspect, the present application provides a device for detecting the remaining amount of a roll of material, the device comprising:

[0061] Image acquisition module, used to collect the image of the material roll during unwinding;

[0062] An extraction module, used for extracting the roll image to obtain a detection image including the roll edge area;

[0063] A judgment module, used to judge whether the current material roll meets the preset conditions based on the feature values ​​in the detection image;

[0064] The output module is used to output a warning message prompting to replace the material roll when the judgment result of the judgment module is yes.

[0065] The material roll remainder detection device provided by the present application has an image acquisition module that acquires the material roll image during unwinding, and an extraction module that extracts a detection image including the edge area of ​​the material roll from the continuous roll image. Then, the judgment module determines whether the current material roll meets a preset condition based on a feature value in the detection image. When the judgment result of the judgment module is yes, the output module outputs a warning message prompting the replacement of the material roll, and the current material roll status can be identified and determined quickly and accurately, with high detection accuracy and low requirements for hardware deployment.

[0066] In a third aspect, the present application provides a material roll remaining amount detection device, comprising: one or more processors; and a memory associated with the one or more processors, the memory being used to store program instructions, which, when read and executed by the one or more processors, executes a material roll remaining amount detection method as provided in any one of the first aspects.

[0067] The material roll remaining amount detection device provided in the present application executes the material roll remaining amount detection method provided in the first aspect, and can quickly and accurately identify and judge the current material roll status, with high detection accuracy and low requirements for hardware deployment.

[0068] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the method provided in any one of the first aspects when the computer program is executed by a processor.

[0069] The computer-readable storage medium provided in the present application implements the material roll remainder detection method provided in the first aspect when the computer program stored thereon is executed by a processor, and can quickly and accurately identify and judge the current material roll status, with high detection accuracy and low requirements for hardware deployment.

[0070] In a fifth aspect, the present application provides a string welding machine, comprising: a photovoltaic cell substrate feeding device, a welding ribbon feeding device, a string conveying device, a string connection device, and a material roll remaining amount detection device as provided in the first aspect;

[0071] The photovoltaic cell substrate feeding device is used to provide the photovoltaic cell substrate to the serial conveying device;

[0072] The welding tape feeding device is used to carry and provide welding tape to the serial connection conveying device. The welding tape feeding device and the photovoltaic cell substrate loading device stack the welding tape and the cell on the serial connection conveying device. The serial connection conveying device welding tape conveying device transports the stacked welding tape and the cell to the serial connection device. The serial connection device connects the stacked welding tape and the cell in series into a cell string.

[0073] The material coil remainder detection device is used to detect the remainder of the material coil in the solder tape feeding device, and is used to warn when the solder tape remainder on the material coil on the solder tape feeding device reaches a critical value.

[0074] The string welder provided in the present application adopts the material coil remaining detection device provided in the first aspect to perform real-time detection of the material coil in the solder tape feeding device that provides the solder tape to the welding conveying device. It can accurately and efficiently detect the material coil remaining in the solder tape feeding device, and output a first warning message prompting replacement of the material coil when the material coil is in a material coil exhaustion state. It can issue a warning prompt when the solder tape on the material coil is about to run out without manual adjustment, thereby reducing the material coil replacement time of the solder tape feeding device, thereby improving the string welding efficiency of the string welder. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] The disclosure of the present application will become more easily understood with reference to the accompanying drawings. It is easy for those skilled in the art to understand that these drawings are only for illustrative purposes and are not intended to limit the scope of protection of the present application. In addition, similar numbers in the drawings are used to represent similar components, among which:

[0076] Figure 1 This is a flow chart of a method for detecting the remaining amount of a roll of material in one embodiment of the present application;

[0077] Figure 2 is a detection flow chart of a method for detecting the remaining amount of a roll of material in one embodiment of the present application;

[0078] Figure 3 It is a structural diagram of a roll remainder detection device according to one of the embodiments of the present application. DETAILED DESCRIPTION

[0079] Some embodiments of the present application are described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are only used to explain the technical principles of the present application and are not intended to limit the protection scope of the present application.

[0080] As described in the background technology, currently, a proximity switch is installed at the position of each material coil on the string welding machine to detect the exhaustion of the welding ribbon. The proximity switch can detect the metal welding wire on the material coil. When the welding wire is exhausted, the proximity switch cannot detect the welding wire and will alarm. However, in this detection method, each material coil needs to be equipped with a proximity switch. Now, each string welding machine needs to use up to dozens of proximity switches, which is costly; the detection range is small, and full-range detection cannot be achieved. When the welding ribbon is almost exhausted, manual adjustment is required; the detection range is single, and it can only detect the exhaustion of metal materials.

[0081] In order to solve the above problems, the present application creatively proposes a method, device, equipment, storage medium and string welding machine for detecting the remaining amount of a material coil. After obtaining the image of the material coil when the solder strip material coil is unrolled, the features in the feature image containing the material coil area in the material coil image are extracted to obtain feature data. According to the feature value, it is judged whether the current state of the material coil is a full reel or an exhausted reel. When the material coil is in the exhausted state, a first warning message prompting the replacement of the material coil is output. Without manual adjustment, an early warning prompt can be given when the solder strip on the material coil is about to run out.

[0082] The present application will be described in detail below through specific embodiments.

[0083] The present application provides a method for detecting the remaining amount of a roll, referring to Figure 1 As shown, the method includes:

[0084] S1. Collect the image of the material roll during unwinding.

[0085] Exemplarily, the image of the material roll during unloading is captured by video. Video capture is implemented by a small industrial color camera, which has a 1080P high-definition resolution and is powered by a POE network. This application does not specifically limit the placement of the camera, as long as it can clearly capture the image of the material roll during unloading. The camera captures the video of the material roll during unloading, and performs frame extraction processing on the video of the material roll during unloading to obtain the image of the material roll during unloading. Exemplarily, refer to Figure 2 As shown, after setting the alarm point in the system in advance, video acquisition is started, and frames of the acquired video are extracted to obtain the image of the material roll during unwinding.

[0086] S2. Extract a feature image including the material roll area from the material roll image.

[0087] S3. Extract the features of the feature image and obtain feature values.

[0088] S4. Determine the state of the current material roll based on the characteristic value. The material roll state at least includes whether the material roll is full or exhausted.

[0089] When the material roll is in the full state, the process continues to execute step S1; when the material roll is in the exhausted state, the process proceeds to S5.

[0090] S5. Output the first warning information for prompting to replace the material roll.

[0091] The method for detecting the remaining amount of a material coil provided by the present application, after obtaining the material coil image when the solder strip material coil is unwinding, extracts features from a feature image containing the material coil area in the material coil image to obtain feature data, and judges whether the current material coil status is a full coil or an exhausted coil according to the feature value, and outputs a first warning message prompting replacement of the material coil when the material coil is in an exhausted coil status, and can issue a warning prompt when the solder strip on the material coil is about to run out without manual adjustment.

[0092] In some embodiments, the material roll status further includes a material roll about to run out status, and the method further includes: when the material roll is in the material roll about to run out status, entering S6 or S7.

[0093] S6. Output a second warning message indicating that the material roll is about to run out, and continue to execute the step of collecting the image of the material roll during unwinding.

[0094] S7, outputting a second warning message indicating that the material roll is about to run out.

[0095] When feature data is obtained based on features in a feature image that includes a material roll area and it is determined based on the feature value that the current material roll is about to run out, a second warning message indicating that the material roll is about to run out is output to issue a warning, and the step of continuing to collect images of the material roll during unwinding is performed to perform continuous real-time detection of the current material roll, or a direct warning is issued and then real-time detection of the replaced current material roll is performed after replacement.

[0096] In some embodiments, step S3 includes:

[0097] S31, performing noise reduction and grayscale processing on the feature image.

[0098] For example, a Gaussian filtering algorithm is used to process noise points on the acquired feature image to reduce image distortion, distortion, blur and other factors that affect the quality and visualization of the image, so as to reduce the interference of external factors such as pad movement. Further remove external lighting factors such as light intensity and color, perform image grayscale processing on the image, reduce the color information of the image, and make the brightness information of the solder strip in the feature image more prominent.

[0099] S32, extracting features of the feature image after noise reduction and grayscale processing, and calculating feature values ​​according to the extracted features.

[0100] Specifically, the features of the feature image after noise reduction and grayscale processing are extracted, and the feature values, namely, RGB values, are calculated according to the extracted features.

[0101] Before extracting the features of the feature image and calculating the feature values, the feature image is subjected to noise reduction and grayscale processing to improve the quality of the feature image and reduce interference with feature extraction.

[0102] In some embodiments, step S32 includes:

[0103] S321. Obtain a grayscale value of each pixel in a first area unit from a characteristic image, and obtain a first characteristic value according to the grayscale value of each pixel in the first area unit.

[0104] Specifically, the first area is the end area of ​​the outlet end of the material roll.

[0105] S322, judging the state of the current material roll based on the characteristic value, including:

[0106] If the first characteristic value belongs to the first value range, it is determined that the material roll is in a full reel state;

[0107] If the first characteristic value belongs to the second value range, it is determined that the material roll is in a material roll exhaustion state.

[0108] Specifically, since the solder strip is white, the RGB value of the white pixel is (255, 255, 255), and the solder pad is black, the RGB value of the black pixel is (0, 0, 0). The first characteristic value is obtained by the gray value of each pixel in the first area unit, that is, the RGB value. If the first characteristic value is within the first value range, it means that the first area unit is white, and it is determined that the material roll is in a full state; if the first characteristic value is within the second value range, it means that the first area unit is black, and it is determined that the material roll is in an exhausted state.

[0109] The grayscale value of each pixel in the first area unit, i.e., at one end of the material roll, is obtained from the feature image, and the first feature value is calculated based on the grayscale value of the pixel. The state of the current material roll is judged based on the first feature value at the end of the material roll and a preset value range, so that the state of the material roll can be detected quickly and accurately.

[0110] Reference Figure 2 As shown, after extracting the frame to obtain the roll feature image, image classification is performed. Image classification is achieved by determining the roll state in the current feature image using the first feature value obtained after executing S31 and S32.

[0111] The alarm point pre-set in the system is the second value range, and the alarm is triggered when the first characteristic value falls within the second value range. Figure 2 As shown, when the image classification result determines that the current reel is full, video acquisition continues; when the image classification result determines that the current solder strip is exhausted, the first eigenvalue belongs to the second value range to trigger an alarm, and a reel change alarm is output.

[0112] In some embodiments, step S32 includes:

[0113] S32A. Obtain the grayscale value of each pixel in the first area unit from the characteristic image, and obtain the first characteristic value according to the grayscale value of each pixel in the first area unit.

[0114] Exemplarily, all pixels in the first area unit are obtained, and the sum of the RGB values ​​of all pixels is calculated. For example, the RGB values ​​of three pixels are (255, 255, 255), (0, 0, 0) and (255, 255, 255), respectively. Then the sum of the three RGB values, that is, the first eigenvalue, is calculated to be 255+255+255+0+0+0+255+255+255=1530.

[0115] Alternatively, all pixels in the first area unit are obtained, and the RGB average values ​​of all pixels are calculated.

[0116] S32B, obtaining the grayscale value of each pixel in the second area unit from the characteristic image, and obtaining the second characteristic value according to the grayscale value of each pixel in the second area unit, and the first area unit and the second area unit are arranged along the axial direction of the roll.

[0117] To facilitate the extraction of feature images, the first area and the second area can be both located at one end of the roll and arranged along the axial direction of the roll; or the first area and the second area can be both located in the middle of the roll and arranged along the axial direction of the roll; of course, the first area and the second area can be the two end portions of the roll respectively.

[0118] S32C, judging the state of the current material roll based on the characteristic value, including:

[0119] If both the first characteristic value and the second characteristic value belong to the first value range, it is determined that the material roll is in a full reel state.

[0120] The first eigenvalue and the second eigenvalue both belong to the first value range, indicating that the first region and the second region are filled with white solder wires, both ends of the coil are covered with solder wires with no black solder pads exposed, and the coil is in a full coil state.

[0121] If both the first characteristic value and the second characteristic value belong to the second value range, it is determined that the roll is in a roll-out state, and the first value range and the second value range have different ranges. In one implementation, the minimum value of the first value range is greater than the maximum value of the second value range.

[0122] The first eigenvalue and the second eigenvalue both belong to the second value range, indicating that both the first region and the second region contain black pads, both ends of the coil have run out of wires and the pads are exposed, and the coil is in an exhausted state.

[0123] By calculating the characteristic values ​​of the areas at both ends of the roll according to the gray values ​​at both ends of the roll, it is determined whether there are welding lines at both ends of the roll according to the characteristic values ​​of the areas at both ends of the roll, thereby more accurately determining the state of the current roll.

[0124] Reference Figure 2As shown, when the image classification result determines that the current reel is in a full state, video acquisition continues to be performed. When the image classification result determines that the current solder tape is exhausted, a reel change alarm is output.

[0125] In some embodiments, the roll state further includes a state where the roll is about to run out, the second area unit is located between the first edge of the roll and the extraction area of ​​the first area unit, and step S32C further includes:

[0126] When the first characteristic value is in the first value range and the second characteristic value is in the second value range, it is determined that the material roll is exhausted in the direction away from the first edge of the material roll; when the first characteristic value is in the first value range and the second characteristic value is in the second value range, that is, there is a white welding wire in the first area, and the welding wire in the second area is exhausted and only the welding pad exists, which means that the welding wire of the first edge of the material roll and the second area in the middle has been exhausted, and the welding wire is consumed in the direction away from the first edge and toward the first area.

[0127] When the first characteristic value is within the second value range and the second characteristic value is within the first value range, it is determined that the material roll is exhausted in a direction toward the first edge of the material roll.

[0128] When the first characteristic value is in the second value range and the second characteristic value is in the first value range, that is, the welding wire is exhausted and only the welding pad exists in the first area, and there is white welding wire in the second area, it means that the welding wire in the first area and the second area in the middle of the roll has been exhausted, and the welding wire is consumed in the direction toward the first edge and away from the first area.

[0129] or,

[0130] The extraction area of ​​the first area unit is located between the first edge of the roll and the extraction area of ​​the second area unit, and the state of the current roll is judged based on the characteristic value, further comprising:

[0131] When the first characteristic value is within the first value range and the second characteristic value is within the second value range, it is determined that the material roll is exhausted in the direction toward the first edge of the material roll; that is, there is a white solder wire in the middle first area, a black solder pad is exposed in the second area at the end, and the solder wire is consumed from the second area toward the first edge.

[0132] When the first characteristic value is in the second value range and the second characteristic value is in the first value range, it is determined that the material roll is exhausted in the direction away from the first edge of the material roll; that is, a black solder pad is exposed in the middle first area, and a white solder wire exists in the second area at the end, and the solder wire is consumed from the first edge at the other end toward the second area.

[0133] According to the value range of the point features in the frame in the image and the point features outside the frame in the image, the material usage of the solder strip inside and outside the frame in the image can be determined, so that the direction in which the solder strip is used up can be accurately determined.

[0134] In some embodiments, after determining that the web is exhausted in a direction toward the first edge of the web, the method further comprises:

[0135] If the characteristic image is close to the first edge of the material roll, it is determined that the material roll is in a state of being exhausted; if the characteristic image is far away from the first edge of the material roll, it is determined that the material roll is in a state of being about to be exhausted;

[0136] After determining that the roll is exhausted in a direction away from the first edge of the roll, the method further includes:

[0137] If the characteristic image is close to the first edge of the material roll, it is determined that the material roll is in a state of being about to run out; if the characteristic image is far away from the first edge of the material roll, it is determined that the material roll is in a state of being run out.

[0138] After determining the direction of the current roll being used up, the specific state of the roll is determined based on whether the characteristic image is close to the first edge of the roll or far from the first edge of the roll, so that the result of determining the state of the roll is more accurate.

[0139] At the same time, the remaining percentage of the roll can also be determined based on the distance between the edge or center of the characteristic image and the edge of the roll. For example, if the distance between the center of the characteristic image and the first edge of the roll is one-fifth of the total length of the roll, and the roll is used up in the direction toward the first edge, it is determined that 20% of the roll is left; if the distance between the center of the characteristic image and the first edge of the roll is four-fifths of the total length of the roll, and the roll is used up in the direction toward the first edge, it is determined that 80% of the roll is left.

[0140] In one implementation, the pad exhaustion alarm can be triggered when 20% of the material roll is exhausted, and when 80% of the material roll is exhausted, the bottom is exposed and no alarm is triggered, and video acquisition can be continued.

[0141] Reference Figure 2 As shown, when the image classification result determines that the pattern content is a solder strip + a solder pad, it is determined that the bottom of the current material roll has been exposed and no alarm is given, and video acquisition continues.

[0142] In some embodiments, the method for detecting the remaining amount of the roll further includes:

[0143] S32D. Obtain the grayscale value of each pixel in the third area unit outside the characteristic image along the axial direction of the material roll from the material roll image, and obtain the third characteristic value according to the grayscale value of each pixel in the third area unit.

[0144] S32E, the third area unit is located between the first edge of the material roll and the characteristic image. When the material roll is used up in the direction away from the first edge of the material roll, the third characteristic value is in the first value range, and the first characteristic value and the second characteristic value are in the second value range, an error is reported; when the material roll is used up in the direction toward the first edge of the material roll, the third characteristic value is in the second value range, and the first characteristic value and the second characteristic value are in the first value range, an error is reported.

[0145] The third area unit is located between the first edge of the roll and the characteristic image. When the roll runs out in the direction away from the first edge of the roll, the welding wire in the third area should run out before the welding wire in the characteristic image. Therefore, an error is reported when the third characteristic value is in the first value range and the first characteristic value and the second characteristic value are in the second value range. When the roll runs out in the direction toward the first edge of the roll, the welding wire in the third area should run out after the welding wire in the characteristic image. Therefore, an error is reported when the third characteristic value is in the second value range and the first characteristic value and the second characteristic value are in the first value range.

[0146] or,

[0147] The characteristic image is located between the first edge of the material roll and the third area unit. When the material roll is used up in the direction toward the first edge of the material roll, the third characteristic value is in the first value range, and the first characteristic value and the second characteristic value are in the second value range, an error is reported; when the material roll is used up in the direction away from the first edge of the material roll, the third characteristic value is in the second value range, and the first characteristic value and the second characteristic value are in the first value range, an error is reported.

[0148] The characteristic image is located between the first edge of the roll and the third area unit. When the roll runs out in the direction toward the first edge of the roll, the welding wire in the third area unit should run out before the welding wire in the characteristic image. Therefore, an error is reported when the third characteristic value is in the first value range and the first characteristic value and the second characteristic value are in the second value range. When the roll runs out in the direction away from the first edge of the roll, the welding wire in the third area unit should run out after the welding wire in the characteristic image. Therefore, an error is reported when the third characteristic value is in the second value range and the first characteristic value and the second characteristic value are in the first value range.

[0149] When it is judged that the solder ribbon inside the image frame is used up but the solder ribbon outside the frame is not used up based on the value range of the point features inside the image frame and the point features outside the image frame, an error reporting process is performed to achieve correction of the material roll remainder detection and avoid erroneous detection affecting production efficiency.

[0150] In some embodiments, reference Figure 2 As shown, after obtaining the image of the material roll during unwinding and before performing image classification, image preprocessing is performed. Specifically, after step S2, the method further includes:

[0151] SⅠ. Use edge detection to detect the edges of objects in the feature image, segment the feature image to obtain a segmented image, and extract the features of the segmented image.

[0152] Specifically, the preset coordinates and the image edge segmentation information are combined to obtain the static image information of the preset identification points on the pad to obtain a segmented image, and the image features are extracted from the segmented image.

[0153] SⅡ. Input the features of the segmented image into a predetermined matching model, in which a reference image including the material roll area is pre-stored.

[0154] Specifically, a feature model of the roll is established by analyzing the features of a large number of rolls of the same specification in advance, which can be implemented based on statistical methods such as mean, variance, etc. The established feature model of the roll is used as a predetermined matching model. The features of the segmented image are input into the predetermined matching model, and the predetermined matching model compares the features of the segmented image with the features of the pre-stored roll. If the difference is large, it is determined to be an abnormal object.

[0155] If the predetermined matching model determines that the segmented image matches the reference image successfully, then continue to execute step S3;

[0156] If the predetermined matching model determines that the segmented image and the reference image are not matched successfully, an error message is given.

[0157] Combined with the preset coordinates and image edge segmentation information, a static image of the preset identification points on the pad is segmented and obtained; a predetermined matching model is used to determine whether the segmented static image can be successfully matched with the reference image, thereby judging whether the segmented static image can meet the detection requirements, avoiding invalid detection of images that do not meet the requirements and reducing detection efficiency.

[0158] The present application also provides a device for detecting the remaining amount of a roll, referring to Figure 3 As shown, the material roll remaining amount detection device includes:

[0159] An image acquisition module 310 is used to acquire images of the material roll during unwinding;

[0160] An extraction module 320 is used to extract the roll image to obtain a detection image including the roll edge area;

[0161] A judgment module 330, for judging whether the current roll meets a preset condition based on the feature value in the detection image;

[0162] The output module 340 is used to output a warning message prompting to replace the material roll when the judgment result of the judgment module is yes.

[0163] In some embodiments, the roll state also includes a roll-out state. When the roll is in the roll-out state, the output module 340 is further used to output a second warning message indicating that the roll is about to run out, and the image acquisition module 310 continues to execute the step of acquiring the roll image during unwinding; or

[0164] The output module 340 is further configured to output a second warning message indicating that the material roll is about to run out.

[0165] In some embodiments, the extraction module 320 includes:

[0166] A processing unit, used for performing noise reduction and grayscale processing on the feature image;

[0167] The extraction and calculation unit is used to extract the features of the feature image after noise reduction and grayscale processing, and calculate the feature value according to the extracted features.

[0168] Optionally, the extraction calculation unit includes:

[0169] An extraction subunit, used to obtain the gray value of each pixel in the first area unit from the feature image, and obtain the first feature value according to the gray value of each pixel in the first area unit;

[0170] The judgment subunit is used to judge the status of the current roll based on the characteristic value.

[0171] If the first characteristic value belongs to the first value range, the judging subunit determines that the material roll is in a full reel state;

[0172] If the first characteristic value belongs to the second value range, the judging subunit determines that the material roll is in a material roll exhaustion state.

[0173] Optionally, the extraction calculation unit includes:

[0174] A first acquisition subunit, used to acquire the gray value of each pixel in the first area unit from the feature image, and obtain a first feature value according to the gray value of each pixel in the first area unit;

[0175] A second acquisition subunit is used to acquire the gray value of each pixel in the second area unit from the characteristic image, and obtain the second characteristic value according to the gray value of each pixel in the second area unit, and the first area unit and the second area unit are arranged along the axial direction of the roll;

[0176] The judgment subunit is used to:

[0177] If both the first characteristic value and the second characteristic value belong to the first value range, it is determined that the material roll is in a full reel state;

[0178] If the first characteristic value and the second characteristic value both belong to the second value range, it is determined that the material roll is in a material roll exhaustion state, and the ranges of the first value range and the second value range are different.

[0179] Optionally, the material roll state also includes a material roll about to run out state, the second area unit is located between the first edge of the material roll and the extraction area of ​​the first area unit, and the judgment subunit is further used for:

[0180] When the first characteristic value is within the first value range and the second characteristic value is within the second value range, it is determined that the roll is exhausted in a direction away from the first edge of the roll;

[0181] When the first characteristic value is within the second value range and the second characteristic value is within the first value range, it is determined that the material roll is exhausted in a direction toward the first edge of the material roll;

[0182] or,

[0183] The extraction area of ​​the first area unit is located between the first edge of the roll and the extraction area of ​​the second area unit, and the judgment subunit is further used for:

[0184] When the first characteristic value is within the first value range and the second characteristic value is within the second value range, it is determined that the roll is exhausted in a direction toward the first edge of the roll;

[0185] When the first characteristic value is within the second value range and the second characteristic value is within the first value range, it is determined that the material roll is exhausted in a direction away from the first edge of the material roll.

[0186] Optionally, the device further comprises:

[0187] a first determination module, configured to determine that the material roll is in a state of being exhausted if the characteristic image is close to the first edge of the material roll after the determination subunit determines that the material roll is exhausted in the direction toward the first edge of the material roll; and to determine that the material roll is in a state of being about to be exhausted if the characteristic image is far from the first edge of the material roll;

[0188] The device also includes:

[0189] The second judgment module is used to judge that the material roll is in a state of being about to run out if the characteristic image is close to the first edge of the material roll after the judgment subunit determines that the material roll is used up in the direction away from the first edge of the material roll; and judge that the material roll is in a state of being run out if the characteristic image is far away from the first edge of the material roll.

[0190] Optionally, the device further comprises:

[0191] An acquisition module, used to acquire the gray value of each pixel point in the third area unit outside the characteristic image along the axial direction of the material roll from the material roll image, and obtain a third characteristic value according to the gray value of each pixel point in the third area unit;

[0192] The third area unit is located between the first edge of the roll and the characteristic image. The output module 340 is further used to report an error when the roll runs out in a direction away from the first edge of the roll, the third characteristic value is in the first value range, and the first characteristic value and the second characteristic value are in the second value range; the output module 340 is further used to report an error when the roll runs out in a direction toward the first edge of the roll, the third characteristic value is in the second value range, and the first characteristic value and the second characteristic value are in the first value range;

[0193] or,

[0194] The characteristic image is located between the first edge of the material roll and the third area unit. The output module 340 is also used to report an error when the material roll is used up in the direction toward the first edge of the material roll, the third characteristic value is in the first value range, and the first characteristic value and the second characteristic value are in the second value range; the output module 340 is also used to report an error when the material roll is used up in the direction away from the first edge of the material roll, the third characteristic value is in the second value range, and the first characteristic value and the second characteristic value are in the first value range.

[0195] In some embodiments, the apparatus further comprises:

[0196] The segmentation and extraction module is used for the extraction module 320 to detect the edge of the object in the feature image by using edge detection after intercepting the feature image including the material roll area from the material roll image, and to segment the feature image to obtain a segmented image, and to extract the features of the segmented image;

[0197] An input matching module is used to input the features of the segmented image into a predetermined matching model, in which a reference image including a roll area is pre-stored;

[0198] If the predetermined matching model in the input matching module determines that the segmented image matches the reference image successfully, the extraction module 320 continues to perform the steps of extracting the features of the feature image and obtaining feature values;

[0199] If the predetermined matching model in the input matching module determines that the segmented image and the reference image are not matched successfully, the output module 340 will report an error.

[0200] The material roll remainder detection device provided by the present application has an image acquisition module that acquires the material roll image during unwinding, and an extraction module that extracts a detection image including the edge area of ​​the material roll from the continuous roll image. Then, the judgment module determines whether the current material roll meets a preset condition based on a feature value in the detection image. When the judgment result of the judgment module is yes, the output module outputs a warning message prompting the replacement of the material roll, and the current material roll status can be identified and determined quickly and accurately, with high detection accuracy and low requirements for hardware deployment.

[0201] The embodiment of the present application also provides a roll remainder detection device, comprising: one or more processors; and a memory associated with the one or more processors, the memory being used to store program instructions, and when the program instructions are read and executed by the one or more processors, performing the following operations:

[0202] Collect the image of the material roll when unwinding;

[0203] Extracting a feature image including a roll area from the roll image;

[0204] Extract the features of the feature image and obtain the feature value;

[0205] Based on the characteristic value, the state of the current material roll is judged, and the material roll state at least includes the material roll being full or the material roll being exhausted;

[0206] When the material roll is in a full reel state, the step of collecting the image of the material roll during unwinding is continued;

[0207] When the material roll is in a state of being exhausted, a first warning message for prompting replacement of the material roll is output.

[0208] When program instructions are read and executed by one or more processors, they also perform the following operations:

[0209] When the material roll is in a state of being about to run out, a second warning message indicating that the material roll is about to run out is output, and the step of collecting the image of the material roll during unwinding is continued; or,

[0210] When the material roll is in a state where the material roll is about to run out, a second warning message indicating that the material roll is about to run out is output.

[0211] When program instructions are read and executed by one or more processors, they also perform the following operations:

[0212] Perform noise reduction and grayscale processing on the feature image;

[0213] The features of the feature image after noise reduction and grayscale processing are extracted, and the feature values ​​are calculated based on the extracted features.

[0214] When program instructions are read and executed by one or more processors, they also perform the following operations:

[0215] Acquire the grayscale value of each pixel in the first area unit from the feature image, and obtain the first feature value according to the grayscale value of each pixel in the first area unit;

[0216] The status of the current roll is determined based on the characteristic values, including:

[0217] If the first characteristic value belongs to the first value range, it is determined that the material roll is in a full reel state;

[0218] If the first characteristic value belongs to the second value range, it is determined that the material roll is in a material roll exhaustion state.

[0219] When program instructions are read and executed by one or more processors, they also perform the following operations:

[0220] Acquire the grayscale value of each pixel in the first area unit from the feature image, and obtain the first feature value according to the grayscale value of each pixel in the first area unit;

[0221] Acquire the gray value of each pixel point in the second area unit from the characteristic image, and obtain the second characteristic value according to the gray value of each pixel point in the second area unit, wherein the first area unit and the second area unit are arranged along the axial direction of the roll;

[0222] The status of the current roll is determined based on the characteristic values, including:

[0223] If the first characteristic value and the second characteristic value both belong to the first value range, it is determined that the material roll is in a full reel state;

[0224] If the first characteristic value and the second characteristic value both belong to the second value range, it is determined that the material roll is in a material roll exhaustion state, and the ranges of the first value range and the second value range are different.

[0225] When program instructions are read and executed by one or more processors, they also perform the following operations:

[0226] When the first characteristic value is within the first value range and the second characteristic value is within the second value range, it is determined that the roll is exhausted in a direction away from the first edge of the roll;

[0227] When the first characteristic value is within the second value range and the second characteristic value is within the first value range, it is determined that the material roll is exhausted in a direction toward the first edge of the material roll;

[0228] or,

[0229] The extraction area of ​​the first area unit is located between the first edge of the roll and the extraction area of ​​the second area unit, and the state of the current roll is judged based on the characteristic value, further comprising:

[0230] When the first characteristic value is within the first value range and the second characteristic value is within the second value range, it is determined that the roll is exhausted in a direction toward the first edge of the roll;

[0231] When the first characteristic value is within the second value range and the second characteristic value is within the first value range, it is determined that the material roll is exhausted in a direction away from the first edge of the material roll.

[0232] When program instructions are read and executed by one or more processors, they also perform the following operations:

[0233] If the characteristic image is close to the first edge of the material roll, it is determined that the material roll is in a state of being exhausted; if the characteristic image is far away from the first edge of the material roll, it is determined that the material roll is in a state of being about to be exhausted;

[0234] After determining that the roll is exhausted in a direction away from the first edge of the roll, the method further includes:

[0235] If the characteristic image is close to the first edge of the material roll, it is determined that the material roll is in a state of being about to run out; if the characteristic image is far away from the first edge of the material roll, it is determined that the material roll is in a state of being run out.

[0236] When program instructions are read and executed by one or more processors, they also perform the following operations:

[0237] Acquire the grayscale value of each pixel point in the third area unit outside the characteristic image along the axial direction of the material roll from the material roll image, and obtain the third characteristic value according to the grayscale value of each pixel point in the third area unit;

[0238] The third area unit is located between the first edge of the roll and the characteristic image. When the roll runs out in the direction away from the first edge of the roll, the third characteristic value is in the first value range, and the first characteristic value and the second characteristic value are in the second value range, an error is reported; when the roll runs out in the direction toward the first edge of the roll, the third characteristic value is in the second value range, and the first characteristic value and the second characteristic value are in the first value range, an error is reported;

[0239] or,

[0240] The characteristic image is located between the first edge of the material roll and the third area unit. When the material roll is used up in the direction toward the first edge of the material roll, the third characteristic value is in the first value range, and the first characteristic value and the second characteristic value are in the second value range, an error is reported; when the material roll is used up in the direction away from the first edge of the material roll, the third characteristic value is in the second value range, and the first characteristic value and the second characteristic value are in the first value range, an error is reported.

[0241] When program instructions are read and executed by one or more processors, they also perform the following operations:

[0242] Use edge detection to detect the edges of objects in the feature image, segment the feature image to obtain a segmented image, and extract the features of the segmented image;

[0243] Inputting the features of the segmented image into a predetermined matching model, wherein a reference image including a roll area is pre-stored in the predetermined matching model;

[0244] If the predetermined matching model determines that the segmented image matches the reference image successfully, then continue to perform the step of extracting features of the feature image and obtaining feature values;

[0245] If the predetermined matching model determines that the segmented image and the reference image are not matched successfully, an error message is given.

[0246] The material roll remaining amount detection device provided in the present application executes the material roll remaining amount detection method provided in the first aspect, and can quickly and accurately identify and judge the current material roll status, with high detection accuracy and low requirements for hardware deployment.

[0247] The present application also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the following operations are performed:

[0248] Collect the image of the material roll when unwinding;

[0249] Extracting a feature image including a roll area from the roll image;

[0250] Extract the features of the feature image and obtain the feature value;

[0251] Based on the characteristic value, the state of the current material roll is judged, and the material roll state at least includes the material roll being full or the material roll being exhausted;

[0252] When the material roll is in a full reel state, the step of collecting the image of the material roll during unwinding is continued;

[0253] When the material roll is in a state of being exhausted, a first warning message for prompting replacement of the material roll is output.

[0254] When the computer program is executed by the processor, the following operations are also performed:

[0255] When the material roll is in a state of being about to run out, a second warning message indicating that the material roll is about to run out is output, and the step of collecting the image of the material roll during unwinding is continued; or,

[0256] When the material roll is in a state where the material roll is about to run out, a second warning message indicating that the material roll is about to run out is output.

[0257] When the computer program is executed by the processor, the following operations are also performed:

[0258] Perform noise reduction and grayscale processing on the feature image;

[0259] The features of the feature image after noise reduction and grayscale processing are extracted, and the feature values ​​are calculated based on the extracted features.

[0260] When the computer program is executed by the processor, the following operations are also performed:

[0261] Acquire the grayscale value of each pixel in the first area unit from the feature image, and obtain the first feature value according to the grayscale value of each pixel in the first area unit;

[0262] The status of the current roll is determined based on the characteristic values, including:

[0263] If the first characteristic value belongs to the first value range, it is determined that the material roll is in a full reel state;

[0264] If the first characteristic value belongs to the second value range, it is determined that the material roll is in a material roll exhaustion state.

[0265] When the computer program is executed by the processor, the following operations are also performed:

[0266] Acquire the grayscale value of each pixel in the first area unit from the feature image, and obtain the first feature value according to the grayscale value of each pixel in the first area unit;

[0267] Acquire the gray value of each pixel point in the second area unit from the characteristic image, and obtain the second characteristic value according to the gray value of each pixel point in the second area unit, wherein the first area unit and the second area unit are arranged along the axial direction of the roll;

[0268] The status of the current roll is determined based on the characteristic values, including:

[0269] If the first characteristic value and the second characteristic value both belong to the first value range, it is determined that the material roll is in a full reel state;

[0270] If the first characteristic value and the second characteristic value both belong to the second value range, it is determined that the material roll is in a material roll exhaustion state, and the ranges of the first value range and the second value range are different.

[0271] When the computer program is executed by the processor, the following operations are also performed:

[0272] When the first characteristic value is within the first value range and the second characteristic value is within the second value range, it is determined that the roll is exhausted in a direction away from the first edge of the roll;

[0273] When the first characteristic value is within the second value range and the second characteristic value is within the first value range, it is determined that the material roll is exhausted in a direction toward the first edge of the material roll;

[0274] or,

[0275] The extraction area of ​​the first area unit is located between the first edge of the roll and the extraction area of ​​the second area unit, and the state of the current roll is judged based on the characteristic value, further comprising:

[0276] When the first characteristic value is within the first value range and the second characteristic value is within the second value range, it is determined that the roll is exhausted in a direction toward the first edge of the roll;

[0277] When the first characteristic value is within the second value range and the second characteristic value is within the first value range, it is determined that the material roll is exhausted in a direction away from the first edge of the material roll.

[0278] When the computer program is executed by the processor, the following operations are also performed:

[0279] If the characteristic image is close to the first edge of the material roll, it is determined that the material roll is in a state of being exhausted; if the characteristic image is far away from the first edge of the material roll, it is determined that the material roll is in a state of being about to be exhausted;

[0280] After determining that the roll is exhausted in a direction away from the first edge of the roll, the method further includes:

[0281] If the characteristic image is close to the first edge of the material roll, it is determined that the material roll is in a state of being about to run out; if the characteristic image is far away from the first edge of the material roll, it is determined that the material roll is in a state of being run out.

[0282] When the computer program is executed by the processor, the following operations are also performed:

[0283] Acquire the grayscale value of each pixel point in the third area unit outside the characteristic image along the axial direction of the material roll from the material roll image, and obtain the third characteristic value according to the grayscale value of each pixel point in the third area unit;

[0284] The third area unit is located between the first edge of the roll and the characteristic image. When the roll runs out in the direction away from the first edge of the roll, the third characteristic value is in the first value range, and the first characteristic value and the second characteristic value are in the second value range, an error is reported; when the roll runs out in the direction toward the first edge of the roll, the third characteristic value is in the second value range, and the first characteristic value and the second characteristic value are in the first value range, an error is reported;

[0285] or,

[0286] The characteristic image is located between the first edge of the material roll and the third area unit. When the material roll is used up in the direction toward the first edge of the material roll, the third characteristic value is in the first value range, and the first characteristic value and the second characteristic value are in the second value range, an error is reported; when the material roll is used up in the direction away from the first edge of the material roll, the third characteristic value is in the second value range, and the first characteristic value and the second characteristic value are in the first value range, an error is reported.

[0287] When the computer program is executed by the processor, the following operations are also performed:

[0288] Use edge detection to detect the edges of objects in the feature image, segment the feature image to obtain a segmented image, and extract the features of the segmented image;

[0289] Inputting the features of the segmented image into a predetermined matching model, wherein a reference image including a roll area is pre-stored in the predetermined matching model;

[0290] If the predetermined matching model determines that the segmented image matches the reference image successfully, then continue to perform the step of extracting features of the feature image and obtaining feature values;

[0291] If the predetermined matching model determines that the segmented image and the reference image are not matched successfully, an error message is given.

[0292] The computer-readable storage medium provided in the present application implements the material roll remainder detection method provided in the first aspect when the computer program stored thereon is executed by a processor, and can quickly and accurately identify and judge the current material roll status, with high detection accuracy and low requirements for hardware deployment.

[0293] The embodiment of the present application also provides a string welding machine, comprising: a photovoltaic cell substrate feeding device, a welding ribbon feeding device, a string conveying device, a string connection device, and a material roll remaining amount detection device as provided in the above embodiment;

[0294] The photovoltaic cell substrate feeding device is used to provide the photovoltaic cell substrate to the serial conveying device;

[0295] The welding tape feeding device is used to carry and provide welding tape to the serial connection conveying device. The welding tape feeding device and the photovoltaic cell substrate loading device stack the welding tape and the cell on the serial connection conveying device. The serial connection conveying device welding tape conveying device transports the stacked welding tape and the cell to the serial connection device. The serial connection device connects the stacked welding tape and the cell in series into a cell string.

[0296] The material coil remainder detection device is used to detect the remainder of the material coil in the solder tape feeding device, and is used to warn when the solder tape remainder on the material coil on the solder tape feeding device reaches a critical value.

[0297] The string welder provided in the present application adopts the material coil remaining detection device provided in the first aspect to perform real-time detection of the material coil in the solder tape feeding device that provides the solder tape to the welding conveying device. It can accurately and efficiently detect the material coil remaining in the solder tape feeding device, and output a first warning message prompting replacement of the material coil when the material coil is in a material coil exhaustion state. It can issue a warning prompt when the solder tape on the material coil is about to run out without manual adjustment, thereby reducing the material coil replacement time of the solder tape feeding device, thereby improving the string welding efficiency of the string welder.

[0298] 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 application. 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.

[0299] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0300] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A method for detecting the remaining amount of a roll, characterized in that: The method comprises: Collect the image of the material roll when unwinding; Extracting a feature image including a roll area from the roll image; Extracting features of the feature image and obtaining feature values; Determine the current state of the material roll based on the characteristic value, the material roll state at least including the material roll being full or the material roll being exhausted; When the material roll is in a full reel state, the step of collecting the image of the material roll during unwinding is continued; When the material roll is in a state of being exhausted, a first warning message for prompting to replace the material roll is output.

2. The method for detecting the remaining amount of a coil according to claim 1, characterized in that: The material roll state also includes a material roll about to run out state, and the method further includes: When the material roll is in a state of being about to run out, a second warning message indicating that the material roll is about to run out is output, and the step of collecting the image of the material roll during unwinding is continued; or, When the material roll is in a state of being about to run out, a second warning message indicating that the material roll is about to run out is output.

3. The method for detecting the remaining amount of a coil according to claim 1, characterized in that: The step of extracting the features of the feature image and obtaining feature values ​​comprises: Performing noise reduction and grayscale processing on the feature image; Extract the features of the feature image after noise reduction and grayscale processing, and calculate the feature value according to the extracted features.

4. The method for detecting the remaining amount of a coil according to claim 3, characterized in that: The extracting features of the feature image after noise reduction and grayscale processing, and calculating feature values ​​according to the extracted features, includes: Acquire the gray value of each pixel in the first area unit from the feature image, and obtain the first feature value according to the gray value of each pixel in the first area unit; The determining the current state of the roll based on the characteristic value includes: If the first characteristic value belongs to the first value range, it is determined that the material roll is in a full reel state; If the first characteristic value belongs to the second value range, it is determined that the material roll is in a material roll exhaustion state.

5. The method for detecting the remaining amount of a coil according to claim 3, characterized in that: The extracting features of the feature image after noise reduction and grayscale processing, and calculating feature values ​​according to the extracted features, includes: Acquire the gray value of each pixel in the first area unit from the feature image, and obtain the first feature value according to the gray value of each pixel in the first area unit; Acquire the gray value of each pixel in the second area unit from the characteristic image, and obtain the second characteristic value according to the gray value of each pixel in the second area unit, wherein the first area unit and the second area unit are arranged along the axial direction of the roll; The determining the current state of the roll based on the characteristic value includes: If both the first characteristic value and the second characteristic value belong to the first value range, it is determined that the material roll is in a full reel state; If both the first characteristic value and the second characteristic value belong to the second value range, it is determined that the material roll is in a material roll exhaustion state, and the first value range and the second value range are different.

6. The method for detecting the remaining amount of a roll of material according to claim 5, characterized in that: The material roll status also includes a state where the material roll is about to run out. The second area unit is located between the first edge of the material roll and the extraction area of ​​the first area unit, and the current state of the material roll is judged based on the characteristic value, further comprising: When the first characteristic value is within a first value range and the second characteristic value is within a second value range, it is determined that the material roll is exhausted in a direction away from the first edge of the material roll; When the first characteristic value is within the second value range and the second characteristic value is within the first value range, it is determined that the material roll is exhausted in a direction toward a first edge of the material roll; or, The extraction area of ​​the first area unit is located between the first edge of the roll and the extraction area of ​​the second area unit, and the determining the current state of the roll based on the characteristic value further includes: When the first characteristic value is within a first value range and the second characteristic value is within a second value range, it is determined that the material roll is exhausted in a direction toward a first edge of the material roll; When the first characteristic value is within the second value range and the second characteristic value is within the first value range, it is determined that the material roll is exhausted in a direction away from the first edge of the material roll.

7. The method for detecting the remaining amount of a coil according to claim 6, characterized in that: After determining that the roll is exhausted in a direction toward the first edge of the roll, the method further includes: If the characteristic image is close to the first edge of the material roll, it is determined that the material roll is in a state of running out of material roll; if the characteristic image is far away from the first edge of the material roll, it is determined that the material roll is in a state of being about to run out of material roll; After determining that the roll is exhausted in a direction away from the first edge of the roll, the method further includes: If the characteristic image is close to the first edge of the material roll, it is determined that the material roll is in a state of being about to run out; if the characteristic image is far away from the first edge of the material roll, it is determined that the material roll is in a state of being run out.

8. The method for detecting the remaining amount of a coil according to claim 7, characterized in that: The material roll remainder detection method also includes: Acquire the grayscale value of each pixel point in the third area unit outside the characteristic image along the axial direction of the material roll from the material roll image, and obtain a third characteristic value according to the grayscale value of each pixel point in the third area unit; The third area unit is located between the first edge of the roll and the characteristic image. When the roll runs out in a direction away from the first edge of the roll, the third characteristic value is in the first value range, and the first characteristic value and the second characteristic value are in the second value range, an error is reported; when the roll runs out in a direction toward the first edge of the roll, the third characteristic value is in the second value range, and the first characteristic value and the second characteristic value are in the first value range, an error is reported; or, The characteristic image is located between the first edge of the material roll and the third area unit. When the material roll is used up along the direction toward the first edge of the material roll, the third characteristic value is located in the first value range, and the first characteristic value and the second characteristic value are located in the second value range, an error is reported; when the material roll is used up along the direction away from the first edge of the material roll, the third characteristic value is located in the second value range, and the first characteristic value and the second characteristic value are located in the first value range, an error is reported.

9. The method for detecting the remaining amount of a coil according to claim 1, characterized in that: After extracting the feature image including the roll area from the roll image, the method further includes: Using edge detection to detect the edge of the object in the feature image, segmenting the feature image to obtain a segmented image, and extracting features of the segmented image; Inputting the features of the segmented image into a predetermined matching model, wherein a reference image including a roll area is pre-stored in the predetermined matching model; If the predetermined matching model determines that the segmented image matches the reference image successfully, then continue to perform the step of extracting the features of the feature image and obtaining feature values; If the predetermined matching model determines that the segmented image is not matched successfully with the reference image, an error message is given.

10. A device for detecting the remaining amount of a roll, characterized in that: The material roll remaining amount detection device comprises: Image acquisition module, used to collect the image of the material roll during unwinding; An extraction module, used for extracting the material roll image to obtain a detection image including the material roll edge area; A judgment module, used for judging whether the current material roll meets the preset conditions based on the feature values ​​in the detection image; The output module is used to output warning information prompting to replace the material roll when the judgment result of the judgment module is yes.

11. A material roll remaining amount detection device, characterized in that: include: one or more processors; And a memory associated with the one or more processors, the memory is used to store program instructions, and when the program instructions are read and executed by the one or more processors, the method for detecting the remaining amount of the material roll as described in any one of claims 1-9 is executed.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 9 are implemented.

13. A string welding machine, characterized in that: include: A photovoltaic cell substrate feeding device, a solder strip feeding device, a serial conveying device, a serial connection device, and a coil remainder detection device as described in claims 1-9; The photovoltaic cell substrate feeding device is used to provide photovoltaic cell substrates to the serial conveying device; The welding tape feeding device is used to carry and provide welding tape to the serial connection conveying device. The welding tape feeding device and the photovoltaic cell substrate feeding device stack the welding tape and the cell on the serial connection conveying device. The serial connection conveying device welding tape conveying device conveys the stacked welding tape and the cell to the serial connection device. The serial connection device connects the stacked welding tape and the cell in series into a cell string. The material coil remainder detection device is used to detect the remainder of the material coil in the solder tape feeding device, and is used to warn when the solder tape remainder on the material coil on the solder tape feeding device reaches a critical value.

Images