A machine vision processing method and system based on image processing

Abstract

The present invention discloses a machine vision processing method and system based on image processing, which relates to the technical field of image processing. The system includes an image processing system and an image acquisition device. The image acquisition device includes a support frame, in which a turntable and a U-shaped frame are rotatably arranged. The turntable is located above the product. The two sides of the U-shaped frame have side plate structures extending down to the sides of the product. A flip frame is rotatably installed on the side plate structures. A plurality of groups of first CCD cameras are arranged at the bottom of the turntable, and a plurality of groups of second CCD cameras are arranged on the flip frame. By using the rotation of the turntable to drive the first CCD cameras to generate displacement conversion, the present invention can collect images of other parts of the product. By driving the rotation of the flip frame and controlling the rotation of the U-shaped frame, the second CCD cameras can collect images of large products in all directions and over a large range, greatly improving the image recognition and detection efficiency of large products.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and particularly relates to a machine vision processing method and system based on image processing. Background Art

[0002] Machine vision inspection refers to a technology that takes pictures of an object through a camera and then processes the captured image to detect features in the object, thereby obtaining the quality condition of the object. Machine vision inspection is increasingly widely used, mainly in dimensional measurement and appearance defect detection. Among them, appearance defect detection plays a decisive role in improving the quality of the entire product.

[0003] Due to the limited size of the CCD camera and the limited graphic resolution, when facing small products, only one set of CCD cameras can complete the detection. However, when facing large products, the imaging range of the CCD camera is limited and cannot fully cover the product surface, thus bringing a great obstacle to product detection and recognition. Summary of the Invention

[0004] The purpose of the present invention is to provide a machine vision processing method and system based on image processing to solve the above deficiencies in the prior art.

[0005] To achieve the above purpose, the present invention provides the following technical solution: A machine vision processing system based on image processing, including an image processing system and an image acquisition device. The image acquisition device includes a support frame, in which a turntable and a U-shaped frame are rotatably arranged. The turntable is located above the product. The two sides of the U-shaped frame have side plate structures extending down to the sides of the product. A flipping frame is rotatably installed on the side plate structures. Multiple groups of first CCD cameras are arranged at the bottom of the turntable, and multiple groups of second CCD cameras are arranged on the flipping frame. In actual use, the product is conveyed to the bottom of the device. The first CCD cameras collect images of the top of the product, and the second CCD cameras collect images of the sides of the product. The rotation of the turntable can drive the first CCD cameras to generate displacement conversion, so as to collect images of other parts of the product. By driving the flipping frame to rotate, the position of the fill light can be converted to collect images of other parts of the side of the product. Controlling the rotation of the U-shaped frame can drive the second CCD cameras to reach other sides of the product, thereby enabling all-round and large-range image collection of large products, and decoding and comparison are performed by the vision processing system to determine whether the product is completely qualified, thus greatly improving the image recognition and detection efficiency of large products.

[0006] Preferably, the support frame includes a top frame. Columns are fixedly connected to the bottoms of both ends of the top frame. A suspension shaft is fixedly connected to the top of the turntable. The suspension shaft is rotationally connected to the top frame and passes through the U-shaped frame and is rotationally connected to the U-shaped frame. A motor directly driving the rotation of the suspension shaft and a motor driving the rotation of the U-shaped frame through a gear assembly are provided on the top frame.

[0007] Preferably, a moving frame is connected to the flipping frame through a linear motor. The moving frame is fixedly installed on the movable end of the linear motor. The second CCD camera is installed on the moving frame. Thus, by driving the moving frame to move horizontally with the linear motor, the second CCD camera can be driven to approach or move away from the product, thereby adapting to products of different sizes.

[0008] Preferably, the support frame further includes a guide rail. A sliding seat is slidably installed on the guide rail. A guide post is fixedly installed on the top of the sliding seat. The column is slidably connected to the guide post. A hydraulic cylinder is fixedly installed on the sliding seat. The movable end of the hydraulic cylinder is fixedly connected to the column. Thus, the turntable as a whole can move, and under the drive of the hydraulic cylinder, the heights of the first CCD camera and the flipping frame can also be adjusted, so as to cooperate with inspecting products of different heights.

[0009] Preferably, a plurality of supplementary light lamps are fixedly installed on the bottom of the turntable and the side of the moving frame to supplement light for the product.

[0010] Preferably, a product conveyor belt is arranged below the turntable, and the product conveyor belt is used for conveying products.

[0011] Preferably, the flipping frame is rotationally connected to the U-shaped frame through a short shaft. Transmission shafts are rotatably installed on both sides of the U-shaped frame. The bottom ends of the transmission shafts are in transmission cooperation with the short shaft through bevel gear assemblies. A gear disk is fixedly connected to the outside of the turntable. The top ends of the transmission shafts are meshed with the gear disk through gears. Thus, when controlling the relative rotation of the U-shaped frame and the turntable, the flipping frame can be synchronously driven to rotate, thereby realizing the synchronous rotation drive of the turntable, the U-shaped frame and the flipping frame, and scanning the whole product in the shortest time.

[0012] A machine vision processing method based on image processing includes the following steps:

[0013] Step 1: Transport the product to the detection area below the turntable and position it between two groups of flipping frames;

[0014] Step 2: Start the first CCD camera and the second CCD camera to collect images of the top and side of the product;

[0015] Step 3: Drive the turntable and the flipping frame to rotate, so that the first CCD camera makes one full circle around the top of the product, and the second CCD camera makes one full circle around the side of the product. Synchronously collect images using the first CCD camera and the second CCD camera to ensure that the image capture range covers the entire top and side wall areas of the product. Each time the turntable and the flipping frame rotate by a certain angle, they then rotate in the reverse direction and reset to avoid entanglement of various wires;

[0016] Step 4: Drive the U-shaped frame to rotate so that the flipping frame reaches the undetected side of the product, and repeat Step 3 to perform image collection again;

[0017] Step 5: The image recognition system decodes the images or video information transmitted by the first CCD camera and the second CCD camera, compares them with qualified products, identifies whether there are defects on the surface of the product, and determines the location of the defects based on the image information of the identified defects. The image information includes the position information corresponding to the first CCD camera or the second CCD camera during image collection.

[0018] In the above technical solution, the technical effects and advantages provided by the present invention are as follows:

[0019] In the present invention, by using the first CCD camera to collect images of the top of the product and the second CCD camera to collect images of the side of the product, the rotation of the turntable can drive the first CCD camera to generate displacement conversion, thereby collecting images of other parts of the product. Driving the flipping frame to rotate can drive the second CCD camera to perform position conversion and collect images of other parts of the side of the product. Controlling the rotation of the U-shaped frame can drive the second CCD camera to reach other sides of the product. Furthermore, it is possible to collect images of large products in a full range and a large scale, and the visual processing system decodes and compares them to determine whether the product is completely qualified, thus greatly improving the image recognition and detection efficiency of large products. Description of the Drawings

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the embodiments.

[0021] Figure 1 It is a schematic diagram of the overall structure of the present invention.

[0022] Figure 2 It is a side view of the flipping frame of the present invention.

[0023] Figure 3 It is a bottom view of the turntable of the present invention.

[0024] Figure 4 It is a detection flow chart of the present invention.

[0025] Description of the Reference Numerals:

[0026] 1. Support frame; 11. Top frame; 12. Column; 13. Guide rail; 14. Slide seat; 15. Guide post; 16. Hydraulic cylinder; 2. Turntable; 21. Hanging shaft; 22. Gear disc; 3. U-shaped frame; 31. Transmission shaft; 4. Tipping frame; 41. Short shaft; 42. Moving frame; 43. Linear motor; 5. First CCD camera; 6. Second CCD camera; 7. Fill light; 8. Product conveyor belt. Detailed implementation manner

[0027] Embodiment 1

[0028] The present invention provides a machine vision processing system based on image processing as Figures 1-3 shown, which includes an image processing system and an image acquisition device. The image acquisition device includes a support frame 1. A turntable 2 and a U-shaped frame 3 are rotatably arranged in the support frame 1. The turntable 2 is located above the product. The two sides of the U-shaped frame 3 have side plate structures extending down to the sides of the product. A tipping frame 4 is rotatably installed on the side plate structures. A plurality of groups of first CCD cameras 5 are arranged at the bottom of the turntable 2. A plurality of groups of second CCD cameras 6 are arranged on the tipping frame 4. During actual use, the product is conveyed to the bottom of the device. The first CCD camera 5 acquires images of the top of the product, and the second CCD camera 6 acquires images of the side of the product. The rotation of the turntable 2 can drive the first CCD camera 5 to generate displacement conversion, so as to acquire images of other parts of the product. By driving the tipping frame 4 to rotate, the fill light 7 can be driven to perform position conversion to acquire images of other parts of the side of the product. And controlling the rotation of the U-shaped frame 3 can drive the second CCD camera 6 to reach other sides of the product, so that images of large products can be acquired in an all-round and large range, and decoded and compared by the vision processing system to determine whether the product is completely qualified, thus greatly improving the image recognition and detection efficiency of large products.

[0029] Further, in the above technical solution, the support frame 1 includes a top frame 11. Columns 12 are fixedly connected to the bottoms of both ends of the top frame 11. A hanging shaft 21 is fixedly connected to the top of the turntable 2. The hanging shaft 21 is rotatably connected to the top frame 11. The hanging shaft 21 passes through the U-shaped frame 3 and is rotatably connected to the U-shaped frame 3. A motor directly driving the hanging shaft 21 to rotate and a motor driving the U-shaped frame 3 to rotate through a gear assembly are arranged on the top frame 11.

[0030] Further, in the above technical solution, a moving frame 42 is connected to the tipping frame 4 through a linear motor 43. The moving frame 42 is fixedly installed on the movable end of the linear motor 43. The second CCD camera 6 is installed on the moving frame 42. Thus, by driving the moving frame 42 to move horizontally with the linear motor 43, the second CCD camera 6 can be driven to approach or move away from the product, so as to adapt to products of different sizes.

[0031] Further, in the above technical solution, the support frame 1 further includes a guide rail 13. A sliding seat 14 is slidably mounted on the guide rail 13. A guide post 15 is fixedly mounted on the top of the sliding seat 14. The column 12 is slidably connected to the guide post 15. A hydraulic cylinder 16 is fixedly mounted on the sliding seat 14. The movable end of the hydraulic cylinder 16 is fixedly connected to the column 12. Thus, the whole turntable 2 can move, and under the drive of the hydraulic cylinder 16, the heights of the first CCD camera 5 and the flipping frame 4 can also be adjusted, so as to cooperate with the inspection of products at different heights.

[0032] Further, in the above technical solution, a plurality of fill light lamps 7 are fixedly mounted on the bottom of the turntable 2 and the side surface of the moving frame 42 to supplement light for the products.

[0033] Further, in the above technical solution, a product conveyor belt 8 is arranged below the turntable 2, and the product conveyor belt 8 is used for conveying products.

[0034] Further, in the above technical solution, the flipping frame 4 is rotationally connected to the U-shaped frame 3 through a short shaft 41. Transmission shafts 31 are rotatably mounted on both sides of the U-shaped frame 3. The bottom end of the transmission shaft 31 is in transmission cooperation with the short shaft 41 through a bevel gear assembly. A gear disk 22 is fixedly connected to the outside of the turntable 2. The top end of the transmission shaft 31 is meshed with the gear disk 22 through a gear 32. Thus, when controlling the relative rotation of the U-shaped frame 3 and the turntable 2, the flipping frame 4 can be synchronously driven to rotate, so as to realize the synchronous rotation drive of the turntable 2, the U-shaped frame 3 and the flipping frame 4, and perform all-round scanning of the products in the shortest time.

[0035] Working principle: In actual use, the product is conveyed to the bottom of the device. The first CCD camera 5 collects images of the top of the product, and the second CCD camera 6 collects images of the side of the product. The rotation of the turntable 2 can drive the first CCD camera 5 to generate displacement conversion, so as to collect images of other parts of the product. Driving the flipping frame 4 to rotate can drive the second CCD camera 6 to perform position conversion and collect images of other parts of the side of the product. Controlling the rotation of the U-shaped frame 3 can drive the second CCD camera 6 to reach other sides of the product. Thus, images of large products can be collected in an all-round and large range, and decoded and compared by the vision processing system to determine whether the product is completely qualified, thereby greatly improving the image recognition and detection efficiency of large products.

[0036] Embodiment 2

[0037] This embodiment provides a machine vision processing method based on image processing, including the following steps:

[0038] Step 1: Transport the product to the detection area below the turntable 2 and locate it between the two flipping frames 4;

[0039] Step 2: Start the first CCD camera 5 and the second CCD camera 6 to collect images of the top and side of the product;

[0040] Step 3: Drive the turntable 2 and the flipping frame 4 to rotate, so that the first CCD camera 5 makes one full circle around the top of the product, and the second CCD camera 6 makes one full circle around the side of the product. Synchronously use the first CCD camera 5 and the second CCD camera 6 to collect images, ensuring that the image capture range covers the entire top and side wall areas of the product. Each time the turntable 2 and the flipping frame 4 rotate a certain angle, they rotate in the reverse direction and reset to avoid entanglement of various wires;

[0041] Step 4: Drive the U-shaped frame 3 to rotate so that the flipping frame 4 reaches the undetected side of the product, and repeat Step 3 to collect images again;

[0042] Step 5: The image recognition system decodes the images or video information transmitted by the first CCD camera 5 and the second CCD camera 6, compares them with qualified products, identifies whether there are defects on the surface of the product, and judges the location of the defects based on the image information of the identified defects. The image information includes the position information corresponding to the first CCD camera 5 or the second CCD camera 6 during image capture.

Claims

1. A machine vision processing system based on image processing, comprising an image processing system and an image acquisition device, characterized in that: The image acquisition device includes a support frame (1). A turntable (2) and a U-shaped frame (3) are rotatably arranged in the support frame (1). The turntable (2) is located above the product. The two sides of the U-shaped frame (3) have side plate structures extending downward to the sides of the product. A turning frame (4) is rotatably installed on the side plate structures. Multiple first CCD cameras (5) are arranged at the bottom of the turntable (2). Multiple second CCD cameras (6) are arranged on the turning frame (4). The support frame (1) includes a top frame (11). Columns (12) are fixedly connected to the bottoms of both ends of the top frame (11). A suspension shaft (21) is fixedly connected to the top of the turntable (2). The suspension shaft (21) is rotatably connected to the top frame (11). The suspension shaft (21) passes through the U-shaped frame (3) and is rotatably connected to the U-shaped frame (3). A motor directly driving the suspension shaft (21) to rotate and a motor driving the U-shaped frame (3) to rotate through a gear assembly are arranged on the top frame 11. The turning frame (4) is rotatably connected to the U-shaped frame (3) through a short shaft (41). Transmission shafts (31) are rotatably installed on both sides of the U-shaped frame (3). The bottom ends of the transmission shafts (31) are in transmission cooperation with the short shaft (41) through bevel gear assemblies. A gear disc (22) is fixedly connected to the outside of the turntable (2). The top ends of the transmission shafts (31) are meshed with the gear disc (22) through gears (32).

2. The machine vision processing system based on image processing according to claim 1, characterized in that: A moving frame (42) is connected to the turning frame (4) through a linear motor (43). The moving frame (42) is fixedly installed on the moving end of the linear motor (43). The second CCD camera (6) is installed on the moving frame (42).

3. A machine vision processing system based on image processing according to claim 2, characterized in that: The support frame (1) further includes a guide rail (13). A sliding seat (14) is slidably installed on the guide rail (13). A guide post (15) is fixedly installed on the top of the sliding seat (14). The column (12) is slidably connected to the guide post (15). A hydraulic cylinder (16) is fixedly installed on the sliding seat (14). The moving end of the hydraulic cylinder (16) is fixedly connected to the column (12).

4. A machine vision processing system based on image processing according to claim 3, characterized in that: Multiple supplementary light lamps (7) are fixedly installed on the bottom of the turntable (2) and the side of the moving frame (42).

5. A machine vision processing system based on image processing according to claim 1, wherein: A product conveyor belt (8) is arranged below the turntable (2). The product conveyor belt (8) is used to convey products.

6. A processing method of the machine vision processing system based on image processing as described in claim 1, characterized in that, It includes the following steps: Step 1: Transport the product to the detection area below the turntable (2) and locate it between the two turning frames (4). Step 2: Start the first CCD camera (5) and the second CCD camera (6) to collect images of the top and side of the product. Step 3: Drive the turntable (2) and the turning frame (4) to rotate, so that the first CCD camera (5) makes a full circle around the top of the product, and the second CCD camera (6) makes a full circle around the side of the product. Synchronously use the first CCD camera (5) and the second CCD camera (6) to collect images to ensure that the image capture range covers the entire top and side wall areas of the product. Step 4: Drive the U-shaped frame (3) to rotate so that the turning frame (4) reaches the undetected side of the product, and repeat Step 3 to collect images again. Step Five: The image recognition system decodes the image or video information transmitted by the first CCD camera (5) and the second CCD camera (6), compares it with the qualified products, identifies whether there are defects on the product surface, and determines the location of the defects based on the image information of the identified defects. The image information includes the position information corresponding to the first CCD camera (5) or the second CCD camera (6) during image acquisition.

Images