Detecting, weighing, sorting and identifying system and method of automatic weighing machine
By introducing a detection weighing sorting recognition system in the automatic weighing machine, using cameras and deep learning algorithms to identify the shape and density of objects, and selecting suitable weighing sensors, the accuracy reduction and equipment of traditional automatic weighing machines at the extreme weight are solved. The damage problem achieves higher weighing accuracy and reliability.
Patent Information
- Application Number
- CN202411079336.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-07
- Publication Date
- 2024-12-03
AI Technical Summary
The load cell of traditional automatic weighing machines has reduced accuracy when handling the ultimate weight, which may cause damage or inability to weigh, affecting normal operation.
The detection weighing and sorting recognition system is adopted, including a weighing module, a camera recognition module, a volume recognition module and a weight estimation module, which captures images through the camera, uses deep learning algorithms to identify the shape and density of objects, and selects a suitable weighing sensor for weighing. , avoid range mismatch.
The weighing accuracy and reliability of the automatic weighing machine are improved, and the accuracy reduction and equipment damage caused by range mismatch in traditional methods are avoided.
Smart Images

Figure CN119063828A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of automatic weighing machines, and in particular to a detection, weighing, sorting and identifying system and method for an automatic weighing machine. Background Art
[0002] The automatic weighing machine is an online weight sorting device that relies on the dynamic resolution capability provided by high-precision weighing sensors. It can remove products that do not meet the qualified weight range in accordance with regulatory requirements, and can also grade products according to weight grade requirements. Through integrated weighing technology, it can achieve fast and accurate sorting.
[0003] The emergence of automatic weighing machines has made the weighing speed extremely fast, which is much more efficient than manual weighing. It can quickly handle the weighing and classification of large quantities of goods, thereby greatly improving production efficiency. Using automatic weighing machines on the production line can complete the work faster than manual work, reducing working time and labor costs. Automatic weighing machines use high-precision digital technology to accurately measure the weight of goods, ensuring that the size and weight of each shipping unit are the same. Compared with manual weighing, automatic weighing machines can avoid the influence of human factors on weighing, making the finished product more accurate and standardized. The emergence of automatic weighing machines can automate weighing and other tasks, reducing manual processing time and labor costs. There is no need for manual participation in weighing work. The machine can complete multiple tasks at the same time, thereby reducing the company's labor costs.
[0004] However, traditional automatic weighing machines have the following disadvantages:
[0005] The weighing sensors installed in traditional automatic weighing machines need to weigh and sort products within a specified range when used. If the product weight is close to the upper and lower limits of the weighing sensor, the weighing accuracy will be reduced, and if the product is too heavy, the weighing sensor may be damaged by gravity. If the product is too light and below the lower sensing limit of the weighing sensor, the automatic weighing machine will be unable to weigh the product, affecting the normal operation and accurate detection of the automatic weighing machine. Summary of the invention
[0006] The purpose of the present invention is to provide a detection, weighing, sorting and identification system and method for an automatic weighing machine, so as to solve the problem proposed in the above background technology that the weighing sensor installed in the traditional automatic weighing machine needs to weigh and sort the products within a specified range when used. If the product is too heavy, the weighing sensor may be damaged by gravity compression. If the product is too light and below the sensing lower limit of the weighing sensor, the automatic weighing machine will be unable to weigh the weight of the product, affecting the normal operation of the automatic weighing machine.
[0007] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: a detection, weighing, sorting and identifying system and method for an automatic weighing machine, comprising a weighing, sorting and identifying system body, wherein the weighing, sorting and identifying system body comprises a weighing module, a camera identification module, a volume identification module, a weight estimation module and a control module, wherein the weighing module, the camera identification module, the volume identification module and the weight estimation module are all connected to the control module;
[0008] The weighing module is an automatic weighing machine, the camera recognition module identifies the appearance, material and environment of the product, the volume recognition module calculates the volume according to the product image transmitted by the camera recognition module, the weight estimation module estimates the weight of the product according to the product volume calculated by the volume recognition module and compares and analyzes it with the weighing module, the control module connects the various modules together, each module sends the demand to the control module, and the control module transmits the demand signal to the module that solves the demand signal.
[0009] As a preferred technical solution of the present invention, the camera recognition module includes a printer, a camera, an alarm, a humidity sensor, a temperature sensor, a display screen and a control unit. The printer, camera, alarm, humidity sensor, temperature sensor and display screen are all connected to the control unit. The printer prints the product image taken by the camera in black and white, color or 3D and stores the printed content. The camera records the product in real time. The alarm alarms the product that exceeds the camera's camera range. The humidity sensor senses the humidity of the weighing environment in real time. The temperature sensor senses the temperature of the weighing environment in real time. The display screen displays the image of the photographed product in real time. The control unit controls the camera's shooting time, shooting angle, and the sensing time of the humidity sensor and the temperature sensor.
[0010] As a preferred technical solution of the present invention, the volume recognition module includes a memory, an image recognition unit, a polyhedron volume calculation unit, an irregular volume calculation unit, an image segmentation unit, a cube calculator, a rectangular parallelepiped calculator, a triangular prism calculator, a pyramid calculator, a prism calculator and a cylinder calculator, the cube calculator, the rectangular parallelepiped calculator, the triangular prism calculator, the pyramid calculator, the prism calculator and the cylinder calculator are all connected to the polyhedron volume calculation unit, the image segmentation unit is connected to the irregular volume calculation unit, the polyhedron volume calculation unit, the irregular volume calculation unit and the memory are all connected to the image recognition unit, the image recognition unit performs shape recognition on the image content recognized by the camera recognition module, and the polyhedron volume calculation unit recognizes the shape of the image content recognized by the camera recognition module. The image is calculated, the irregular volume calculation unit calculates the irregular graphics according to the recognition, the graphics segmentation unit cuts the irregular graphics into multiple regular images, and the memory stores the identified images for subsequent volume comparison. The calculation formula of the cube calculator is, a is the side length of the cube, the calculation formula of the rectangular parallelepiped calculator is, a, b, h are all side lengths, the calculation formula of the triangular prism calculator is, F is the base area, h is the height, the calculation formula of the pyramid calculator is, F is the base area, h is the height, the calculation formula of the prism calculator is, F1, F2 are the areas of two parallel bases respectively, and h is the height of the ground distance. The calculation formula of the cylinder calculator is, R is the outer radius, h is the height, and r is the inner radius of the hollow straight cylinder.
[0011] As a preferred technical solution of the present invention, the weight estimation module includes a size comparison unit, a density estimation unit and a volume processing unit. The size comparison unit compares the volume calculated by the volume recognition module and the volume estimated by the volume processing unit and confirms the final estimated volume. The volume processing unit uses a deep learning algorithm to identify the shape, texture and features of different objects through a training model, and determines whether they meet the preset standards before calculating the volume. The density estimation unit predicts the density range based on the image recognized by the camera recognition module.
[0012] As a preferred technical solution of the present invention, the weighing module includes several weighing sensors, a data recording unit, a data management unit and a communication unit. The data recording unit records the weighing values of the weighing module, the data management unit classifies and manages the weighing data, and the communication unit transmits the weighing information to the background. Several weighing sensors have different ranges. According to the weight estimated by the weight estimation module, a suitable weighing sensor is connected and the weighing value is determined based on the measured value of the weight sensor and the weight and curve change diagram compared with the estimated value.
[0013] The present invention provides a method for using a detection, weighing, sorting and identifying system of an automatic weighing machine, comprising the following steps:
[0014] Step 1: Image capture: The camera recognition module captures the product image, and the printer prints the product image captured by the camera in black and white, color, or 3D and stores the printed content;
[0015] Step 2, estimate the volume: if the captured image is an irregular shape, cut it into regular images and then calculate the volume of each segment and then sum them up. If the image is a regular shape, use a calculator to calculate the volume. The calculation formula of the cube calculator is: a is the side length of the cube. The calculation formula of the rectangular parallelepiped calculator is: a, b, h are all side lengths. The calculation formula of the triangular prism calculator is: F is the bottom area and h is the height. The calculation formula of the pyramid calculator is: F is the bottom area and h is the height. The calculation formula of the prism calculator is: F1 and F2 are the areas of two parallel bottom surfaces respectively, and h is the height of the distance from the ground. The calculation formula of the cylinder calculator is: R is the outer radius, h is the height, and r is the inner radius. The hollow straight cylinder is;
[0016] Step 3: Estimating weight: The weight of the product is predicted based on the estimated volume and estimated density. The volume processing unit uses a deep learning algorithm to train a model to identify the shape, texture, and features of different objects, and determines whether they meet the preset standards before calculating the volume and estimating the weight through the density estimation unit.
[0017] Step 4. Product weighing: Connect a suitable weighing sensor according to the weight estimated by the weight estimation module and compare the measured value of the weight sensor and the weight and curve change diagram with the estimated value to determine the weighing value.
[0018] Compared with the prior art, the present invention has the following beneficial effects:
[0019] The system captures images of objects to be weighed through cameras. Using high-resolution cameras and light control technology, the system can clearly obtain image information of objects. Next, the volume recognition module processes and analyzes the collected images. It uses deep learning algorithms to train models to identify the shapes, textures and features of different objects, and determines whether they meet the preset standards. At the same time, the module will estimate the weight of the object based on its size and density, and select a weighing sensor with a corresponding range for weighing, to avoid the phenomenon that the weighing sensor range does not match the product weight, thereby ensuring the normal operation of the automatic weighing machine and the weighing detection accuracy. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] Figure 1 It is a schematic diagram of the architecture of the weighing, sorting and identifying system of the present invention;
[0021] Figure 2 Schematic diagram of the architecture of the camera recognition module of the present invention;
[0022] Figure 3 It is a schematic diagram of the architecture of the volume recognition module of the present invention;
[0023] Figure 4 is a schematic diagram of the architecture of the weight estimation module of the present invention;
[0024] Figure 5 It is a schematic diagram of the structure of the weighing module of the present invention;
[0025] Figure 6 It is a flow chart of the present invention. DETAILED DESCRIPTION
[0026] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0027] See also Figure 1-6 The present invention provides a detection, weighing, sorting and identifying system for an automatic weighing machine, including a weighing, sorting and identifying system body, the weighing, sorting and identifying system body including a weighing module, a camera identification module, a volume identification module, a weight estimation module and a control module, the weighing module, the camera identification module, the volume identification module and the weight estimation module are all connected to the control module;
[0028] The weighing module is an automatic weighing machine. The camera recognition module identifies the product's appearance, material and the environment in which the product is located. The volume recognition module calculates the volume based on the product image transmitted by the camera recognition module. The weight estimation module estimates the product's weight based on the product volume calculated by the volume recognition module and compares and analyzes it with the weighing module. The control module connects each module together, and each module sends its demand to the control module. The control module transmits the demand signal to the module that solves the demand signal.
[0029] The camera recognition module includes a printer, a camera, an alarm, a humidity sensor, a temperature sensor, a display screen and a control unit. The printer, camera, alarm, humidity sensor, temperature sensor and display screen are all connected to the control unit. The printer prints the product image taken by the camera in black and white, color or 3D and stores the printed content. The camera records the product in real time. The alarm alarms for products that exceed the camera's camera range. The humidity sensor senses the humidity of the weighing environment in real time. The temperature sensor senses the temperature of the weighing environment in real time. The display screen displays the image of the photographed product in real time. The control unit controls the camera's shooting time, shooting angle, and the sensing time of the humidity sensor and the temperature sensor.
[0030] The volume recognition module includes a memory, an image recognition unit, a polyhedron volume calculation unit, an irregular volume calculation unit, an image segmentation unit, a cube calculator, a rectangular parallelepiped calculator, a triangular prism calculator, a pyramid calculator, a prism calculator and a cylinder calculator. The cube calculator, the rectangular parallelepiped calculator, the triangular prism calculator, the pyramid calculator, the prism calculator and the cylinder calculator are all connected to the polyhedron volume calculation unit. The image segmentation unit is connected to the irregular volume calculation unit. The polyhedron volume calculation unit, the irregular volume calculation unit and the memory are all connected to the image recognition unit. The image recognition unit performs shape recognition on the image content recognized by the camera recognition module. The polyhedron volume calculation unit performs calculation according to the recognized regular image. The irregular volume calculation unit performs irregular graphic calculation according to the recognition. The graphic segmentation unit cuts the irregular graphic into multiple regular images. The memory stores the recognized images for subsequent volume comparison. The calculation formula of the cube calculator is V=a 3 , a is the side length of the cube, the calculation formula of the rectangular parallelepiped calculator is V=a·b·h, a, b, h are all side lengths, the calculation formula of the triangular prism calculator is V=F·h, F is the base area, h is the height, and the calculation formula of the pyramid calculator is F is the base area, h is the height, and the calculation formula of the prism calculator is: F 1 , F 2 are the areas of two parallel bottom surfaces respectively, h is the height of the ground distance. The calculation formula of the cylinder calculator is V = nR 2 h, R is the outer radius, h is the height, r is the inner radius. For a hollow straight cylinder, V = nh (R 2 -r 2 ).
[0031] The weight estimation module includes a size comparison unit, a density estimation unit and a volume processing unit. The size comparison unit compares the volume calculated by the volume recognition module and the volume estimated by the volume processing unit and confirms the final estimated volume. The volume processing unit uses a deep learning algorithm to train a model to identify the shape, texture and features of different objects, and determines whether they meet the preset standards before calculating the volume. The density estimation unit predicts the density range based on the image recognized by the camera recognition module.
[0032] The weighing module includes several weighing sensors, a data recording unit, a data management unit and a communication unit. The data recording unit records the weighing values of the weighing module, the data management unit classifies and manages the weighing data, and the communication unit transmits the weighing information to the background. Several weighing sensors have different ranges. According to the weight estimated by the weight estimation module, a suitable weighing sensor is connected and the weighing value is determined based on the measured value of the weight sensor and the weight and curve change diagram compared with the estimated value.
[0033] The present invention provides a method for using a detection, weighing, sorting and identifying system of an automatic weighing machine, comprising the following steps:
[0034] Step 1: Image capture: The camera recognition module captures the product image, and the printer prints the product image captured by the camera in black and white, color, or 3D and stores the printed content;
[0035] Step 2: Estimate the volume: If the image is irregular, cut it into regular images and calculate the volume of each segment and then sum them up. If the image is regular, calculate the volume using a calculator. The calculation formula of the cube calculator is V = a 3 , a is the side length of the cube, the calculation formula of the rectangular parallelepiped calculator is V=a·b·h, a, b, h are all side lengths, the calculation formula of the triangular prism calculator is V=F·h, F is the base area, h is the height, and the calculation formula of the pyramid calculator is F is the base area, h is the height, and the calculation formula of the prism calculator is: F 1 , F 2 are the areas of two parallel bottom surfaces respectively, h is the height of the ground distance. The calculation formula of the cylinder calculator is V = nR 2 h, R is the outer radius, h is the height, r is the inner radius. For a hollow straight cylinder, V = nh (R 2 -r 2 );
[0036] Step 3: Estimating weight: The weight of the product is predicted based on the estimated volume and estimated density. The volume processing unit uses a deep learning algorithm to train a model to identify the shape, texture, and features of different objects, and determines whether they meet the preset standards before calculating the volume and estimating the weight through the density estimation unit.
[0037] Step 4. Product weighing: Connect a suitable weighing sensor according to the weight estimated by the weight estimation module and compare the measured value of the weight sensor and the weight and curve change diagram with the estimated value to determine the weighing value.
[0038] In the present invention, the camera recognition module captures the image of the product, and the printer prints the product image captured by the camera in black and white, in color or in 3D and stores the printed content; if the captured image is an irregular pattern, it is cut into regular images and then the volume is calculated and summed up in sections; if the image is a regular pattern, the volume is calculated by a calculator, and the calculation formula of the cube calculator is V=a 3 , a is the side length of the cube, the calculation formula of the rectangular parallelepiped calculator is V=a·b·h, a, b, h are all side lengths, the calculation formula of the triangular prism calculator is V=F·h, F is the base area, h is the height, and the calculation formula of the pyramid calculator is F is the base area, h is the height, and the calculation formula of the prism calculator is: F 1 , F 2 are the areas of two parallel bottom surfaces respectively, h is the height of the ground distance. The calculation formula of the cylinder calculator is V = nR 2 h, R is the outer radius, h is the height, r is the inner radius. For a hollow straight cylinder, V = nh (R 2 -r 2 ); predict the weight of the product based on the estimated volume and estimated density, the volume processing unit uses a deep learning algorithm to identify the shape, texture and features of different objects through a training model, and determines whether it meets the preset standards before calculating the volume and cooperating with the density estimation unit to estimate the weight; connect a suitable weighing sensor according to the weight estimated by the weight estimation module and determine the weighing value based on the measured value of the weight sensor and the weight and curve change diagram compared with the estimated value.
[0039] Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments, or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.
Claims
1. A detection, weighing, sorting and identifying system for an automatic weighing machine, comprising a weighing, sorting and identifying system body, characterized in that: The weighing, sorting and identifying system body comprises a weighing module, a camera identification module, a volume identification module, a weight estimation module and a control module, wherein the weighing module, the camera identification module, the volume identification module and the weight estimation module are all connected to the control module; The weighing module is an automatic weighing machine, the camera recognition module identifies the appearance, material and environment of the product, the volume recognition module calculates the volume according to the product image transmitted by the camera recognition module, the weight estimation module estimates the weight of the product according to the product volume calculated by the volume recognition module and compares and analyzes it with the weighing module, the control module connects the various modules together, each module sends the demand to the control module, and the control module transmits the demand signal to the module that solves the demand signal.
2. The detection, weighing, sorting and identification system of an automatic weighing machine according to claim 1, characterized in that: The camera recognition module includes a printer, a camera, an alarm, a humidity sensor, a temperature sensor, a display screen and a control unit. The printer, camera, alarm, humidity sensor, temperature sensor and display screen are all connected to the control unit. The printer prints the product image captured by the camera in black and white, color or 3D and stores the printed content. The camera records the product in real time. The alarm alarms the product that exceeds the camera range. The humidity sensor senses the humidity of the weighing environment in real time. The temperature sensor senses the temperature of the weighing environment in real time. The display screen displays the image of the photographed product in real time. The control unit controls the camera shooting time, shooting angle, humidity sensor and temperature sensor sensing time.
3. The detection, weighing, sorting and identification system of an automatic weighing machine according to claim 1, characterized in that: The volume recognition module includes a memory, an image recognition unit, a polyhedron volume calculation unit, an irregular volume calculation unit, an image segmentation unit, a cube calculator, a rectangular parallelepiped calculator, a triangular prism calculator, a pyramid calculator, a prism calculator and a cylinder calculator. The cube calculator, the rectangular parallelepiped calculator, the triangular prism calculator, the pyramid calculator, the prism calculator and the cylinder calculator are all connected to the polyhedron volume calculation unit. The image segmentation unit is connected to the irregular volume calculation unit. The polyhedron volume calculation unit, the irregular volume calculation unit and the memory are all connected to the image recognition unit. The image recognition unit performs shape recognition on the image content recognized by the camera recognition module. The polyhedron volume calculation unit performs calculation according to the recognized regular image. The irregular volume calculation unit performs irregular graphic calculation according to the recognition. The graphic segmentation unit cuts the irregular graphic into multiple regular images. The memory stores the recognized images for subsequent volume comparison. The calculation formula of the cube calculator is V=a 3 , a is the side length of the cube, the calculation formula of the rectangular parallelepiped calculator is V=a·b·h, a, b, h are all side lengths, the calculation formula of the triangular prism calculator is V=F·h, F is the base area, h is the height, and the calculation formula of the pyramid calculator is F is the base area, h is the height, and the calculation formula of the prism calculator is: F1, F2 are the areas of two parallel bottom surfaces, h is the height of the ground distance. The calculation formula of the cylinder calculator is V = nR 2 h, R is the outer radius, h is the height, r is the inner radius. For a hollow straight cylinder, V = nh (R 2 -r 2 ).
4. The detection, weighing, sorting and identification system of an automatic weighing machine according to claim 1, characterized in that: The weight estimation module includes a size comparison unit, a density estimation unit and a volume processing unit. The size comparison unit compares the volume calculated by the volume recognition module and the volume estimated by the volume processing unit and confirms the final estimated volume. The volume processing unit uses a deep learning algorithm to identify the shape, texture and features of different objects through a training model, and determines whether they meet the preset standards before calculating the volume. The density estimation unit predicts the density range based on the image recognized by the camera recognition module.
5. The detection, weighing, sorting and identification system of an automatic weighing machine according to claim 1, characterized in that: The weighing module includes several weighing sensors, a data recording unit, a data management unit and a communication unit. The data recording unit records the weighing values of the weighing module, the data management unit classifies and manages the weighing data, and the communication unit transmits the weighing information to the background. Several weighing sensors have different ranges. According to the weight estimated by the weight estimation module, a suitable weighing sensor is connected and the weighing value is determined according to the measured value of the weight sensor and the weight and curve change diagram compared with the estimated value.
6. A method for using a detection, weighing, sorting and identifying system for an automatic weighing machine according to any one of claims 1 to 5, characterized in that: The following steps are involved: Step 1: Image capture: The camera recognition module captures the product image, and the printer prints the product image captured by the camera in black and white, color, or 3D and stores the printed content; Step 2: Estimate the volume: If the image is irregular, cut it into regular images and calculate the volume of each segment and then sum them up. If the image is regular, calculate the volume using a calculator. The calculation formula of the cube calculator is V = a 3 , a is the side length of the cube, the calculation formula of the rectangular parallelepiped calculator is V=a·b·h, a, b, h are all side lengths, the calculation formula of the triangular prism calculator is V=F·h, F is the base area, h is the height, and the calculation formula of the pyramid calculator is F is the base area, h is the height, and the calculation formula of the prism calculator is: F1, F2 are the areas of two parallel bottom surfaces, h is the height of the ground distance. The calculation formula of the cylinder calculator is V = nR 2 h, R is the outer radius, h is the height, r is the inner radius. For a hollow straight cylinder, V = nh (R 2 -r 2 ); Step 3: Estimating weight: The weight of the product is predicted based on the estimated volume and estimated density. The volume processing unit uses a deep learning algorithm to train a model to identify the shape, texture, and features of different objects, and determines whether they meet the preset standards before calculating the volume and estimating the weight through the density estimation unit. Step 4. Product weighing: Connect a suitable weighing sensor according to the weight estimated by the weight estimation module and compare the measured value of the weight sensor and the weight and curve change diagram with the estimated value to determine the weighing value.
