Product defect detection device and method based on infrared thermal imaging

Abstract

The invention provides a product defect detection device and method based on infrared thermal imaging, the product defect detection device comprises a machine table, a conveying module, a camera shooting module, an infrared detection module, a sorting module and a control module, the conveying module and the camera shooting module are installed on the machine table, and the conveying module carries a product to be detected to pass through the lower part of the camera shooting module; the infrared detection modules are located on the two sides of the conveying direction of the conveying module and conduct infrared imaging on section burr defects of passing to-be-detected products, the control module controls the camera module to photograph and position the to-be-detected products, and the control module analyzes and detects infrared imaging images. The sorting module is used for sorting defective products analyzed and detected by the control module; through active infrared imaging of the infrared detection module, high-precision burr detection of the cutting section of the aluminum part of the radiator can be realized, the intelligent level of detection can be effectively improved, and through active infrared imaging detection, the sorting accuracy is improved, and the device cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field related to product appearance detection, and in particular to a product defect detection device and detection method based on infrared thermal imaging; it is mainly used for appearance quality detection of aluminum products used in the industrial field after sawing, specifically, aluminum edge burr defect detection based on machine vision and infrared imaging, and according to the appearance characteristics of the aluminum cross section, active infrared imaging detection is performed on its edge, thereby completing real-time detection and sorting of high-quality aluminum parts. Background Art

[0002] Although the radiator is small, it is an important device for dissipating heat from the heat-prone electronic components in electrical appliances. For example, the computer will be heated for a long time, and the radiator in the CPU / GPU can play a cooling role, thereby extending its service life. In particular, high-performance GPUs cannot do without radiators. There are many material options for radiators. For example, silver, copper, aluminum and steel. Different materials have an important influence on the heat dissipation effect of the radiator. From the perspective of thermal conductivity analysis, the best is silver, but silver is not the first material because it is too expensive. The thermal conductivity of copper is close behind, almost twice that of aluminum. Considering the cost-effectiveness, aluminum is the preferred material in the civilian market.

[0003] A common manufacturing solution for aluminum radiators is the extrusion process, which is suitable for mass production of aluminum radiators with low investment, short development cycle and low production cost. For example, the commonly used aluminum extrusion material is aluminum 6063, which has good thermal conductivity. In radiator processing, the aluminum extrusion process is mainly used to manufacture lamellar fins or cylindrical fins. Its working principle is to heat the aluminum ingot to 520-540℃ under high pressure, so that the aluminum liquid flows through the extrusion die with grooves to make the heat sink. Then cut the heat sink blank to form the profile groove and other processing processes.

[0004] However, there is a problem that needs to be solved in the extrusion process of aluminum heat sinks. Since aluminum materials have good ductility, after cutting the heat sink blank, there may be burrs on the processed section. This seriously affects the quality of the product and also affects the later use of the heat sink, such as the movement of airflow, heat dissipation efficiency, etc. Therefore, the general solution is to add a deburring process and process it mechanically. In addition, in order to further improve product quality, a quality inspection process is arranged after the deburring process. At present, the commonly used quality inspection is to use machine vision methods.

[0005] Although computer-based machine vision methods are more common now, they generally use visible light cameras to take pictures, and the technical means used are relatively simple. When the burrs are small, or the shapes and positions are complex, it is impossible to make accurate judgments, and there is a risk of missed detection; or the aluminum is dipped in emulsifier for detection, but the cost and complexity of the detection device are increased, and the detection time is increased. For example, the following invention patent of the People's Republic of China, with the publication number CN118096680A, is named "A defect detection method system and storage medium for fine scratches on the surface of aluminum materials". The backbone network is used to extract the multi-scale hierarchical feature information of the image and output a multi-level feature map; for the feature map, sparse sampling is performed based on the loss cost to generate a sparse sampling map. The sparse sampling map is multiplied with the high-level feature map, and then fused with the low-level feature map modulated by the semantic information of the high-level feature map, the feature block is segmented and input into the encoder to add positioning information; the decoder is used to decode the feature information, and the predicted positioning bounding box and target category (defects on the surface of aluminum materials) are output. Another example is the invention patent of the People's Republic of China with publication number CN117911399A, entitled "A lightweight multi-scale aluminum profile surface defect detection method", which discloses a lightweight multi-scale aluminum profile surface defect detection method. The method adopts a convolution kernel attention module to help adapt the receptive field size, so as to accurately locate the defect target, connect low-level and high-level features by using bidirectional feature fusion, and fuse image features of different sizes to enable the network to have a more comprehensive learning of multi-scale defects. The trained model is quantized with INT8 mixed precision and deployed to the TensorRT platform for acceleration, and the location and type of defects in the analyzed image are obtained, and visualized on the original image. Another example is the invention patent of the People's Republic of China with publication number CN103308660A, entitled "A Method for Detecting Surface Defects of Aviation Aluminum Materials", which detects surface defects of aluminum materials and collects information; heat treats the aluminum materials in a heat treatment furnace according to the detected defect records; dips the heat-treated aluminum materials in an emulsifier, and gently shakes or stirs the aluminum materials during dipping; hot-dip coats the aluminum materials dipped in the emulsifier in a molten bath, thereby forming an alloy coating on the aluminum materials. The method for detecting surface defects of aviation aluminum materials of the present invention can be effectively used for imaging and detecting defects on non-flat surfaces, thereby improving the accuracy of detection.

[0006] Although the above patents propose aluminum defect detection and grading devices, most of them only use traditional visible light machine vision methods; or test the aluminum after dipping it in emulsifier, which increases the cost and complexity of the detection device. For aluminum radiator defect detection, especially burr defects, there is a high probability of missed detection when the size is small and the shape and location are complex. Summary of the invention

[0007] The main purpose of the present invention is to overcome the above shortcomings and deficiencies of the prior art and to provide a product defect detection device and detection method based on infrared thermal imaging.

[0008] A product defect detection device based on infrared thermal imaging comprises a machine platform, a transmission module, a camera module, an infrared detection module, a sorting module and a control module, wherein the transmission module and the camera module are mounted on the machine platform, the transmission module carries a product to be detected and passes through the bottom of the camera module, the infrared detection module is mounted on the machine platform and is located on both sides of the transmission direction of the transmission module, the infrared detection module performs infrared imaging of the cross-sectional burr defects of the product to be detected passing through, the control module controls the camera module to take a picture of the product to be detected and locate it and transmit the picture back to the control module, the control module analyzes and detects the infrared imaging image of the cross-section of the product to be detected, and the sorting module sorts the defective products analyzed and detected by the control module.

[0009] The present invention can realize high-precision burr detection on the cut section of radiator aluminum parts through active infrared imaging of the infrared detection module. On the one hand, it can effectively improve the intelligent level of detection. On the other hand, through active infrared imaging detection, it can improve the sorting accuracy and reduce the cost of the device.

[0010] In one embodiment, the infrared detection module includes a motion guide mechanism, multiple active thermal actuation arrays and an infrared camera module. The motion guide mechanism operates under the control of the control module to provide positioning for the active thermal actuation arrays and the infrared camera module.

[0011] In one embodiment, the motion guide mechanism includes a stepper motor, a screw rod kit, a zero position sensor and a displacement platform, the stepper motor and the zero position sensor are electrically connected to the control module, the zero position sensor is used to detect whether the displacement platform is reset, the stepper motor is dynamically connected to the screw rod kit, the screw rod kit is fixedly connected to the displacement platform, and the active thermal excitation array and the infrared camera module are installed on the displacement platform.

[0012] In one embodiment, the active thermal actuation array includes a plurality of thermal actuation components arranged side by side, and the thermal actuation components are actuated by eddy current inductive heating, thermal radiation heating or hot air heating.

[0013] In one implementation of the embodiment, the infrared camera module is electrically connected to the control module, and the infrared camera module takes a picture of the product cross-section area under the active excitation array under the control of the control module to obtain an infrared imaging image.

[0014] In one implementation of the embodiment, the infrared camera module includes a plurality of infrared cameras, a plurality of protective covers and a cooling pipe; each of the infrared cameras is equipped with a corresponding protective cover, and the cooling pipe surrounds the infrared camera.

[0015] In one embodiment, the conveying module includes a conveyor belt and a plurality of fixing fixtures, wherein the plurality of fixing fixtures are installed on the conveyor belt at a certain interval, and the products to be inspected are installed on the fixing fixtures.

[0016] A product defect detection method based on infrared thermal imaging comprises the following steps:

[0017] Step 1: After the camera module initially takes pictures and positions, the control module controls the infrared detection module to reach the outside of the cross section of the product to be detected on the module to be transmitted;

[0018] Step 2: The control module calculates the positional relationship between the infrared detection module and the cross section of the product to be detected and performs infrared imaging of the cross section of the product to be detected by adjusting the positional relationship;

[0019] Step 3, the infrared detection module sends the infrared imaging image back to the control module;

[0020] Step 4: The control module analyzes the image data of the infrared imaging. If it is determined that there are burrs at the current position of the product section, the detection is stopped, and the control module controls the sorting module to sort the defective products detected by the analysis; otherwise, it goes to step 2 and performs infrared imaging detection on the next position of the product section until all areas are detected.

[0021] In one embodiment, step 4 further includes:

[0022] The infrared imaging image is segmented. The infrared detection module includes a motion guide mechanism, an active thermal excitation array and an infrared camera module, and matches the area within the detection range of each active thermal excitation array internal part;

[0023] The infrared imaging of each segmented image is converted into a corresponding pseudo-color image according to the temperature;

[0024] For each pseudo-color image, the area, perimeter, shape, and center coordinate feature values ​​of the region with a temperature higher than the threshold are counted;

[0025] A fusion algorithm is used to evaluate the area, perimeter, shape, and center coordinates of the region to determine whether the pseudo-color image is the infrared image corresponding to the burr.

[0026] In one implementation of the embodiment, determining in step 4 whether the pseudo-color image is an infrared image corresponding to the burr includes the following steps:

[0027] Collect data and obtain the segmentation pseudo-color image features through experimental design, X1 = area, X2 = perimeter, X3 = shape, X4 = center coordinates, Y = 0 / 1 (0 = no burrs, 1 = burrs), the capacity of the sample library is greater than 200;

[0028] Training model: Through continuous training of the SVM model, the relationship between X1, X2, X3, X4 and Y is established, that is, the coupling correlation model between the pseudo-color image features and the burrs is established;

[0029] Defect detection, during detection, the segmented pseudo-color image features are input into the trained SVM model to obtain the recognition result and store the current recognition result into the control module.

[0030] The present invention has the following beneficial effects: through the active infrared imaging of the infrared detection module, high-precision burr detection on the cut section of the radiator aluminum parts can be achieved. On the one hand, the intelligence level of the detection can be effectively improved. On the other hand, through active infrared imaging detection, the sorting accuracy can be improved and the device cost can be reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a schematic diagram of a product defect detection device based on infrared thermal imaging of the present invention;

[0032] Figure 2 It is a partial structural schematic diagram of a product defect detection device based on infrared thermal imaging of the present invention;

[0033] Figure 3 It is a partial structural schematic diagram of a product defect detection device based on infrared thermal imaging of the present invention;

[0034] Figure 4 It is a schematic structural diagram of another angle of the product defect detection device based on infrared thermal imaging of the present invention;

[0035] Figure 5 It is a schematic diagram of the electric control module of the product defect detection device based on infrared thermal imaging of the present invention;

[0036] Figure 6 It is a schematic diagram of an auxiliary module of the product defect detection device based on infrared thermal imaging of the present invention;

[0037] Figure 7 This is a training model diagram of the product defect detection method based on infrared thermal imaging of the present invention. DETAILED DESCRIPTION

[0038] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0039] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise", "axial", "radial", "circumferential" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0040] 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 one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0041] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0042] In the present invention, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may mean that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, a first feature being "above", "above" or "above" a second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is higher in level than the second feature. A first feature being "below", "below" or "below" a second feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is lower in level than the second feature.

[0043] The present invention provides a product defect detection device based on infrared thermal imaging, comprising a machine platform, a transmission module, a camera module, an infrared detection module, a sorting module and a control module, wherein the transmission module and the camera module are mounted on the machine platform, the transmission module carries a product to be detected and passes through the bottom of the camera module, the infrared detection module is mounted on the machine platform and is located on both sides of the transmission direction of the transmission module, the infrared detection module performs infrared imaging on the cross-section burr defects of the product to be detected that passes through, the control module controls the camera module to take a picture of the product to be detected and locate it and transmit the picture back to the control module, the control module analyzes and detects the infrared imaging image of the cross section of the product to be detected, and the sorting module sorts the defective products analyzed and detected by the control module.

[0044] The present invention can realize high-precision burr detection on the cut section of radiator aluminum parts through active infrared imaging of the infrared detection module. On the one hand, it can effectively improve the intelligent level of detection. On the other hand, through active infrared imaging detection, it can improve the sorting accuracy and reduce the cost of the device.

[0045] Embodiment 1

[0046] See also Figures 1 to 4 The present invention provides a product defect detection device based on infrared thermal imaging, which is mainly used to detect whether there is an aluminum heat sink with burrs. The device comprises a machine 1, a transmission module 2, a camera module 3, an infrared detection module 4, a sorting module 5 and a control module. The transmission module 2 and the camera module 3 are installed on the machine 1, wherein the camera module 3 is installed above the machine 1 by means of a column suspension, the transmission module 2 carries the product to be detected and passes through the bottom of the camera module 3, the infrared detection module 4 is installed on the machine 1 and is located on both sides of the transmission direction of the transmission module 2, the infrared detection module 4 performs infrared imaging on the cross-section burr defects of the product to be detected, the control module controls the camera module 3 to take pictures and locate the product to be detected and transmit the pictures back to the control module, wherein the camera module 3 is a high-resolution camera, the control module analyzes and detects the infrared imaging image of the cross section of the product to be detected, and the sorting module 5 sorts the defective products analyzed and detected by the control module.

[0047] More specifically, see Figure 2 and Figure 3 The infrared detection module 4 includes a motion guide mechanism 41, a plurality of active thermal actuation arrays 42 and an infrared camera module 43. The motion guide mechanism 41 moves under the control of the control module to provide positioning for the active thermal actuation arrays 42 and the infrared camera module 43. The motion track of the motion guide mechanism 41 moves along the conveying direction of the conveying module 2, so as to facilitate the positioning of the active thermal actuation arrays 42 and the infrared camera module 43 during the detection process.

[0048] More specifically, see Figure 2 and Figure 3 The motion guide mechanism 41 includes a stepper motor 411, a screw rod kit 412, a zero position sensor 413 and a displacement platform 414. The stepper motor 411 and the zero position sensor 413 are electrically connected to the control module. The zero position sensor 413 is used to detect whether the displacement platform 414 is reset. The stepper motor 411 is dynamically connected to the screw rod kit 412. The screw rod kit 412 is fixedly connected to the displacement platform 414. The displacement platform 414 is equipped with an active thermal excitation array 42 and an infrared camera module 43.

[0049] More specifically, the active thermal actuation array 42 includes a plurality of thermal actuation components arranged side by side. The active thermal actuation array 42 generally includes 2 to 3 groups of thermal actuation components arranged vertically, that is, perpendicular to the plane of the conveying module 2. Preferably, the number of thermal actuation components is 60 to 90. The thermal actuation components are excited by eddy current inductive heating, thermal radiation heating or hot air heating.

[0050] When the excitation method is eddy current, the active thermal excitation array 42, under the control of the control module, performs inductive heating on the outer edge of the cut section contour line of the aluminum part to be inspected (the area where burrs may appear), that is, the induction coil current of the internal parts of the active thermal excitation array 42 is controlled (medium / high frequency alternating current is applied) to generate a time-varying magnetic field, so that the edge material of the aluminum part section is quickly heated up.

[0051] When the excitation method is thermal radiation, the active thermal excitation array 42, under the control of the control module, performs thermal radiation heating on the outer edge of the cut section contour line of the aluminum part to be inspected (the area where burrs may appear), that is, the heating coil current of the internal parts of the active thermal excitation array 42 is controlled (AC / DC power is applied) to generate a high-temperature thermal radiation source, so that the edge material of the aluminum part section is quickly heated up.

[0052] When the excitation method is hot air, the active thermal excitation array 42, under the control of the control module, outputs controllable hot air to the outer edge of the cut section contour line of the aluminum part to be inspected (the area where burrs may appear), that is, the airflow switch of the internal parts of the active thermal excitation array 42 is controlled, and the hot air of the high-pressure air module is connected to the rear end of the internal parts of the active thermal excitation array 42, so that the edge material of the aluminum part section is quickly heated up.

[0053] More specifically, the infrared camera module 43 is electrically connected to the control module. Under the control of the control module, the infrared camera module 43 takes a picture of the product cross-section area under the active excitation array 42 to obtain an infrared imaging image.

[0054] More specifically, see Figure 2, the infrared camera module 43 includes a plurality of infrared cameras, a plurality of protective covers and cooling pipes; each infrared camera is equipped with a corresponding protective cover, and the cooling pipe surrounds the infrared camera. Avoid the negative impact of excess heat on the infrared camera module 43 under active thermal excitation; further, by configuring the cooling pipe, the peripheral local environment of the infrared camera can also be cooled to reduce the risk of failure of infrared electronic components under heat. Preferably, the infrared camera, the protective cover and the cooling pipe are assembled / installed inside the displacement platform 414 to form an integrated infrared camera module 43. It can be understood that the number of internal components of the active thermal excitation array 42 and the infrared camera module 43 can be increased or decreased according to the requirements of detection speed, cost, etc.

[0055] The present invention can realize high-precision burr detection on the cut section of radiator aluminum parts through active infrared imaging of the infrared detection module. On the one hand, it can effectively improve the intelligent level of detection. On the other hand, through active infrared imaging detection, it can improve the sorting accuracy and reduce the cost of the device.

[0056] More specifically, see Figure 1 The conveying module 2 includes a conveyor belt 21 and a plurality of fixing fixtures 22. The plurality of fixing fixtures 22 are installed on the conveyor belt 21 at a certain interval. The fixing fixtures 22 are provided with the products to be inspected. The infrared detection modules 4 are installed on both sides of the fixing fixtures 22, respectively, for infrared imaging analysis and detection of burr defects on the cross section. Preferably, the fixing fixtures 22 and the aluminum parts to be inspected are used in pairs. The width of the fixing fixtures 22 can be adjusted to accommodate aluminum parts of different specifications and sizes within a certain range, generally with a 10% fluctuation in the width direction, to meet its versatility. As for the aluminum parts to be inspected with obvious differences in specifications, the fixing fixtures 22 need to be remade.

[0057] More specifically, see Figure 4The sorting module 5 includes a vertical column 51, a vertical lifting motor 52, a vertical lifting mechanism 53, a horizontal moving motor 54, a horizontal moving mechanism 55, a clamping mechanism 56, a mobile platform 57 and a waste conveyor belt 58; the vertical column 51 is fastened above the machine platform 1 by bolts, the vertical lifting motor 52 is installed at an upper position on one side of the vertical column 51, the vertical lifting motor 52 is electrically connected to the control module, the power of the vertical lifting motor 52 is matched and connected with the vertical lifting mechanism 53, and the mobile platform 57 is installed on the vertical lifting mechanism 53, so that the movement of the vertical lifting motor 52 can drive the mobile platform 57 moves up and down in the vertical direction; the horizontal moving motor 54, the horizontal moving mechanism 55, and the clamping mechanism 56 are installed on the mobile platform 57; the horizontal moving mechanism 55 is driven by the horizontal moving motor 54 to realize the horizontal movement of the clamping mechanism 56, that is, to move along the moving direction of the waste conveyor belt 58, and the horizontal moving motor 54 is electrically connected to the control module; the clamping mechanism 56 is realized by pneumatic means to complete the grasping / releasing of the aluminum radiator to be inspected, and the pneumatic control of the clamping mechanism 56 is electrically connected through the control module; the sorting module 5, under the control of the control module, grabs the aluminum radiator with burrs out of the normal inspection line. Preferably, a defective product collection box 6 is set below the end of the waste conveyor belt 58 of the sorting module 5 to collect aluminum radiators with burr defects.

[0058] More specifically, see Figure 5 and Figure 6 The control module includes an electric control module 7 and an auxiliary module 8. The electric control module 7 includes an embedded ARM single chip microcomputer 71, an input / output submodule 72, an industrial Ethernet communication submodule 73, and a display submodule 74. The embedded ARM single chip microcomputer 71 is connected to the input / output submodule 72 through an on-chip bus. The auxiliary module 8 includes an industrial Ethernet 81, an annular LED light 82, an alarm 83, a power module 84, and a high-pressure air module 85. The industrial Ethernet 81, the annular LED light 82, and the alarm 83 respectively realize communication, information display, and alarm. The power module 84 provides power for the eddy current and thermal radiation of the active thermal excitation array 42.

[0059] Among them, the embedded ARM single-chip computer 71 is connected to the camera module 3 and the infrared detection module 4 through the industrial Ethernet 81 of the auxiliary module 8; the input / output submodule 72 is respectively connected to the motion guide mechanism 41, the active thermal excitation array 42 and the auxiliary module 8 control signal, and controls the corresponding stepping motor, solenoid valve, and cylinder action according to the control process; the input / output submodule 72 is also connected to and controls the opening and closing of the ring LED light 82 and its brightness; the input / output submodule 72 is also connected to the alarm 83. The power module 84 is electrically connected to the electronic control module 7 through the industrial Ethernet 81; the high-pressure air module 85 provides temperature / pressure controllable air for the hot air of the active thermal excitation array 42, and the high-pressure air module 85 is electrically connected to the electronic control module 7 through the industrial Ethernet 81. It can be understood that the auxiliary module 8 is a common device installed under the machine 1. Among them, the cooling pipe of the infrared camera module 43 is connected to the high-pressure air module 85 of the auxiliary module 8; when working, the high-pressure air module 85 transports 20-25°C air to the cooling pipe.

[0060] Preferably, the eddy current and thermal radiation excitation methods of the active thermal excitation array 42 can quickly heat up very small burrs and burrs on the cross section of the aluminum part, so that the connection strength between the burrs and the cross section of the aluminum part after heating is reduced, and the burrs are removed by hot air from the high-pressure air module 85, so that detection and deburring can be integrated, and the next process can be directly entered, especially burrs and burrs of specific shapes can be quickly removed by thermal radiation, and the subsequent deburring and deburring processes and re-detection processes can also be reduced, which greatly reduces the process flow.

[0061] Embodiment 2

[0062] The present invention provides a product defect detection method based on infrared thermal imaging, comprising the following steps:

[0063] Step 1: After the camera module initially takes pictures and positions, the control module controls the infrared detection module to reach the outside of the cross section of the product to be detected on the module to be transmitted;

[0064] Step 2: The control module calculates the positional relationship between the infrared detection module and the cross section of the product to be detected, and performs infrared imaging of the cross section of the product to be detected by adjusting the positional relationship; wherein the positional relationship includes being far away, inside or intersecting, and the control module only controls the operation of the thermally actuated parts of the infrared detection module that are in the intersection state, and other thermally actuated parts are in a non-operating state;

[0065] Step 3, the infrared detection module sends the infrared imaging image back to the control module;

[0066] Step 4: The control module analyzes the image data of the infrared imaging. If it is determined that there are burrs at the current position of the product section, the detection is stopped, and the control module controls the sorting module to sort the defective products detected by the analysis; otherwise, it goes to step 2 and performs infrared imaging detection on the next position of the product section until all areas are detected.

[0067] In one embodiment, step 4 further includes:

[0068] The infrared imaging image is segmented. The infrared detection module includes a motion guide mechanism, an active thermal excitation array and an infrared camera module, and matches the area within the detection range of each active thermal excitation array internal part;

[0069] The infrared imaging of each segmented image is converted into a corresponding pseudo-color image according to the temperature;

[0070] For each pseudo-color image, the area, perimeter, shape, and center coordinate feature values ​​of the region with a temperature higher than the threshold are counted, where the threshold is 120°C;

[0071] A fusion algorithm is used to evaluate the area, perimeter, shape, and center coordinates of the region to determine whether the pseudo-color image is the infrared image corresponding to the burr.

[0072] See also Figure 7 In one embodiment, the step 4 of determining whether the pseudo-color image is an infrared image corresponding to the burr comprises the following steps:

[0073] Collect data, obtain segmentation pseudo-color image features through experimental design, X1 = area, X2 = perimeter, X3 = shape, X4 = center coordinate, Y = 0 / 1 (0 = no burr, 1 = burr), the capacity of the sample library is greater than 200; preferably, the experimental design is an orthogonal experimental design / CCD experimental design;

[0074] Training model: Through continuous training of the SVM model, the relationship between X1, X2, X3, X4 and Y is established, that is, the coupling correlation model between the pseudo-color image features and the burrs is established;

[0075] Defect detection, during detection, the segmented pseudo-color image features are input into the trained SVM model to obtain the recognition result and store the current recognition result into the control module.

[0076] Among them, in the first implementation mode of this embodiment, when the active thermal excitation array is excited by eddy current, the active thermal excitation array, under the control of the control module, performs inductive heating on the outer edge of the cut section contour line of the aluminum part to be inspected (the area where burrs may appear), that is, the induction coil current of the internal parts of the active thermal excitation array is controlled (applying medium / high frequency alternating current) to generate a time-varying magnetic field, so that the edge material of the aluminum part section is quickly heated up.

[0077] In the second implementation of this embodiment, when the active thermal excitation array is excited by thermal radiation, the active thermal excitation array, under the control of the control module, performs thermal radiation heating on the outer edge of the cut section contour line of the aluminum part to be inspected (the area where burrs may appear), that is, the current of the heating coil of the internal parts of the active thermal excitation array is controlled (AC / DC power is applied) to generate a high-temperature thermal radiation source, thereby rapidly heating the edge material of the aluminum part section.

[0078] In the third implementation of the present embodiment, when the active thermal excitation array is excited by hot air, the active thermal excitation array, under the control of the control module, outputs controllable hot air to the outer edge of the cut section contour line of the aluminum part to be inspected (the area where burrs may appear), that is, the airflow switch of the internal parts of the active thermal excitation array is controlled, and the hot air of the high-pressure air module 85 is connected to the rear end of the internal parts of the active thermal excitation array, so that the edge material of the aluminum part section is quickly heated up.

[0079] Preferably, the eddy current and thermal radiation excitation methods of the active thermal excitation array can quickly heat up very small burrs and burrs on the cross section of the aluminum part, so that the connection strength between the burrs and the cross section of the aluminum part after heating is reduced, and the burrs are removed by hot air from the high-pressure air module, which can realize detection and deburring in one, and can directly enter the next process. In particular, burrs and burrs of specific shapes can be quickly removed by thermal radiation, and the subsequent deburring and deburring processes and re-detection processes can also be reduced, greatly reducing the process flow.

[0080] Among them, the sorting module sorts the aluminum radiators after inspection according to the inspection results of the control module; if there is no burr defect, the sorting module does not perform any action and moves to the next process according to the preset process of the conveying module; if a burr defect is found, the sorting module, under the delayed control of the control module, moves it to the waste conveyor belt through the clamping mechanism and then drops it into the defective product collection box.

[0081] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0082] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A product defect detection device based on infrared thermal imaging, characterized in that: include: A machine, a transmission module, a camera module, an infrared detection module, a sorting module and a control module, wherein the transmission module and the camera module are installed on the machine, the transmission module carries the product to be detected and passes through the bottom of the camera module, the infrared detection module is installed on the machine and is located on both sides of the transmission direction of the transmission module, the infrared detection module performs infrared imaging of the cross-sectional burr defects of the product to be detected passing through, the control module controls the camera module to take pictures and locate the product to be detected and transmit the pictures back to the control module, the control module analyzes and detects the infrared imaging image of the cross-section of the product to be detected, and the sorting module sorts the defective products analyzed and detected by the control module.

2. The product defect detection device based on infrared thermal imaging according to claim 1 is characterized in that: The infrared detection module includes a motion guide mechanism, a plurality of active thermal actuation arrays and an infrared camera module. The motion guide mechanism operates under the control of the control module to provide positioning for the active thermal actuation arrays and the infrared camera module.

3. The product defect detection device based on infrared thermal imaging according to claim 2 is characterized in that: The motion guide mechanism includes a stepper motor, a lead screw kit, a zero position sensor and a displacement platform. The stepper motor and the zero position sensor are electrically connected to the control module. The zero position sensor is used to detect whether the displacement platform is reset. The stepper motor is dynamically connected to the lead screw kit, and the lead screw kit is fixedly connected to the displacement platform. The active thermal excitation array and the infrared camera module are installed on the displacement platform.

4. The product defect detection device based on infrared thermal imaging according to claim 2 is characterized in that: The active thermal actuation array includes a plurality of thermal actuation components arranged side by side, and the actuation mode of the thermal actuation components is eddy current inductive heating, thermal radiation heating or hot air heating.

5. The product defect detection device based on infrared thermal imaging according to claim 2 is characterized in that: The infrared camera module is electrically connected to the control module. Under the control of the control module, the infrared camera module takes pictures of the product cross-section area under the active excitation array to obtain infrared imaging images.

6. The product defect detection device based on infrared thermal imaging according to claim 2 is characterized in that: The infrared camera module comprises a plurality of infrared cameras, a plurality of protective covers and a cooling pipe; each of the infrared cameras is equipped with a corresponding protective cover, and the cooling pipe surrounds the infrared cameras.

7. The product defect detection device based on infrared thermal imaging according to claim 1 is characterized in that: The conveying module comprises a conveying belt and a plurality of fixing fixtures, wherein the plurality of fixing fixtures are installed on the conveying belt at a certain interval, and the products to be inspected are installed on the fixing fixtures.

8. A product defect detection method based on infrared thermal imaging, characterized in that: The steps include: Step 1: After the camera module initially takes pictures and positions, the control module controls the infrared detection module to reach the outside of the cross section of the product to be detected on the module to be transmitted; Step 2: The control module calculates the positional relationship between the infrared detection module and the cross section of the product to be detected and performs infrared imaging of the cross section of the product to be detected by adjusting the positional relationship; Step 3, the infrared detection module sends the infrared imaging image back to the control module; Step 4: The control module analyzes the image data of the infrared imaging. If it is determined that there are burrs at the current position of the product section, the detection is stopped, and the control module controls the sorting module to sort the defective products detected by the analysis; otherwise, it goes to step 2 and performs infrared imaging detection on the next position of the product section until all areas are detected.

9. The product defect detection method based on infrared thermal imaging according to claim 8, characterized in that: The step 4 also includes: The infrared imaging image is segmented. The infrared detection module includes a motion guide mechanism, an active thermal excitation array and an infrared camera module, and matches the area within the detection range of each active thermal excitation array internal part; The infrared imaging of each segmented image is converted into a corresponding pseudo-color image according to the temperature; For each pseudo-color image, the area, perimeter, shape, and center coordinate feature values ​​of the region with a temperature higher than the threshold are counted; A fusion algorithm is used to evaluate the area, perimeter, shape, and center coordinates of the region to determine whether the pseudo-color image is the infrared image corresponding to the burr.

10. The product defect detection method based on infrared thermal imaging according to claim 8, characterized in that: In step 4, judging whether the pseudo-color image is an infrared image corresponding to the burr includes the following steps: Collect data and obtain the segmentation pseudo-color image features through experimental design, X1 = area, X2 = perimeter, X3 = shape, X4 = center coordinates, Y = 0 / 1 (0 = no burrs, 1 = burrs), the capacity of the sample library is greater than 200; Training model: Through continuous training of the SVM model, the relationship between X1, X2, X3, X4 and Y is established, that is, the coupling correlation model between the pseudo-color image features and the burrs is established; Defect detection, during detection, the segmented pseudo-color image features are input into the trained SVM model to obtain the recognition result and store the current recognition result into the control module.

Images