Foreign matter recognition and impurity removal device based on image recognition technology

Abstract

The embodiment of the utility model discloses a foreign matter recognition and impurity removal device based on an image recognition technology, which comprises a first rack, a material thinning mechanism, an image recognition system and a foreign matter removal mechanism, the material thinning mechanism comprises a second rack, a material uniformizing conveying belt, a vibration motor and a spring, and a first driving wheel and a first driven wheel are respectively arranged at the two ends of the second rack; the first driving wheel is connected with an output shaft of the first motor, the vibration motor is arranged below the second rack and fixedly connected with the second rack, the upper end of the spring is connected with the second rack, and the lower end of the spring is connected with the first rack; the foreign matter removing mechanism comprises a third rack and a bidirectional foreign matter discharging conveying belt, the third rack is connected with the first rack, a second driving wheel and a second driven wheel are arranged at the two ends of the third rack respectively, the second driving wheel is connected with an output shaft of a second motor, and the right end of the bidirectional foreign matter discharging conveying belt is arranged below the left end of the material uniformizing conveying belt; the image recognition system comprises a foreign matter detection camera.

Description

Technical Field

[0001] The present application relates to the technical field of tobacco leaf redrying processing, and in particular to a foreign matter recognition and impurity removal device based on image recognition technology. Background Art

[0002] During the leaf threshing and redrying process, the removal of non-tobacco foreign matter in each process section has always been the focus of attention in the industry. According to the different physical properties of foreign matter, different methods are used to remove it from tobacco materials during the processing. However, the ordinary impurity removal mechanism has a poor effect on the removal of small non-metallic foreign matter, and such foreign matter is highly concealed in the moving material and is not easy to be found, which has a great impact on the quality of the finished product. Summary of the invention

[0003] The purpose of the present application is to provide a foreign matter recognition and impurity removal device based on image recognition technology, so as to solve the problem that the ordinary impurity removal mechanism in the prior art has a poor effect in removing small-volume non-metallic foreign matters.

[0004] To achieve the above-mentioned purpose, the embodiment of the present application provides a foreign body identification and impurity removal device based on image recognition technology, comprising: a first frame, a material thinning mechanism, an image recognition system, and a foreign body removal mechanism, wherein:

[0005] The material thinning mechanism includes a second frame, a material leveling conveyor belt, a vibration motor, and a spring. A first driving wheel and a first driven wheel are respectively provided at two ends of the second frame. The first driving wheel is connected to the output shaft of the first motor. The material leveling conveyor belt is sleeved on the first driving wheel and the first driven wheel at two ends of the second frame. The vibration motor is arranged below the second frame and is fixedly connected to the second frame. The upper end of the spring is connected to the second frame, and the lower end is connected to the first frame.

[0006] The foreign body rejection mechanism includes a third frame and a bidirectional foreign body discharge conveyor belt, the third frame is connected to the first frame, and two ends of the third frame are respectively provided with a second driving wheel and a second driven wheel, the second driving wheel is connected to the output shaft of the second motor, and the bidirectional foreign body discharge conveyor belt is sleeved on the second driving wheel and the second driven wheel at two ends of the third frame, and the right end of the bidirectional foreign body discharge conveyor belt is arranged below the left end of the material leveling conveyor belt;

[0007] The image recognition system comprises a foreign object detection camera, which is connected to the first frame and arranged above the left end of the material leveling conveyor belt and / or above the right end of the bidirectional foreign object discharge conveyor belt.

[0008] Optionally, the left end of the material leveling conveyor belt is higher than the right end.

[0009] Optionally, a plurality of vibration plates are provided on the second frame, and the vibration plates extend from one side of the second frame to the other side along the width direction, and the upper and lower side edges of the vibration plates are in contact with the inner surface of the material leveling conveyor belt.

[0010] Optionally, there are at least four springs, which are respectively arranged at the left and right ends of both sides of the second frame.

[0011] Optionally, the foreign matter removal mechanism further includes: a foreign matter selection frame, and the foreign matter selection frame is arranged below the right end of the bidirectional foreign matter discharge conveyor belt.

[0012] Optionally, the image recognition system further includes: a recognition computer and a foreign object image display screen, wherein the recognition computer is connected to the foreign object detection camera and is used to obtain a detection image captured by the foreign object detection camera and display it on the foreign object image display screen.

[0013] Optionally, the foreign object detection camera includes a first camera and a second camera, the first camera is arranged above the left end of the material leveling conveyor belt, and the second camera is arranged above the right end of the bidirectional foreign object discharge conveyor belt.

[0014] Optionally, it also includes: a conveyor belt programmable controller, and the conveyor belt programmable controller is connected to the identification computer, the first motor, and the second motor respectively.

[0015] Optionally, the foreign object detection camera is a JSA-8ELECPTZ3C series foreign object detection camera.

[0016] The embodiments of the present application have the following advantages:

[0017] The embodiment of the present application provides a foreign body recognition and impurity removal device based on image recognition technology, including: a first frame, a material thinning mechanism, an image recognition system, and a foreign body removal mechanism, wherein the material thinning mechanism includes a second frame, a material leveling conveyor belt, a vibration motor, and a spring, and a first driving wheel and a first driven wheel are respectively provided at both ends of the second frame, the first driving wheel is connected to the output shaft of the first motor, the material leveling conveyor belt is sleeved on the first driving wheel and the first driven wheel at both ends of the second frame, the vibration motor is arranged below the second frame and fixedly connected to the second frame, the upper end of the spring is connected to the second frame, and the lower end is connected to the first A frame is connected; the foreign body removal mechanism includes a third frame and a bidirectional foreign body discharge conveyor belt, the third frame is connected to the first frame, and the second driving wheel and the second driven wheel are respectively provided at both ends of the third frame, the second driving wheel is connected to the output shaft of the second motor, and the bidirectional foreign body discharge conveyor belt is sleeved on the second driving wheel and the second driven wheel at both ends of the third frame, and the right end of the bidirectional foreign body discharge conveyor belt is arranged below the left end of the material leveling conveyor belt; the image recognition system includes a foreign body detection camera, the foreign body detection camera is connected to the first frame and is arranged above the left end of the material leveling conveyor belt and / or above the right end of the bidirectional foreign body discharge conveyor belt.

[0018] Compared with the prior art, the foreign matter identification and impurity removal device based on image recognition technology provided by the above technical solution first feeds the thick material into the material thinning mechanism to be vibrated into a loose material thin layer, and then identifies the small-volume non-metallic foreign matter through the image recognition system, and immediately marks the position of the foreign matter, and controls the foreign matter removal mechanism based on the foreign matter position signal to make the bidirectional foreign matter discharge conveyor belt discharge the material containing the foreign matter, thereby greatly improving the rejection rate of small-volume non-metallic foreign matter. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the implementation methods of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for the implementation methods or the description of the prior art. Obviously, the drawings described below are only exemplary, and for ordinary technicians in this field, other implementation drawings can be derived from the provided drawings without creative work.

[0020] Figure 1 A schematic diagram of the three-dimensional structure of a foreign matter recognition and impurity removal device based on image recognition technology provided in at least one embodiment of the present application. DETAILED DESCRIPTION

[0021] The following is a description of the implementation of the present application by specific specific embodiments. People familiar with the technology can easily understand other advantages and effects of the present application from the contents disclosed in this specification. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0022] In the description of the present application, it should be noted that the terms "center", "up", "down", "left", "right", "vertical", "horizontal", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present application. In addition, the terms "first", "second", and "third" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance. Unless otherwise clearly specified and limited, the terms "set", "install", "connected", and "connected" 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 it can be an indirect connection through an intermediate medium, or it can be a connection between the two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific circumstances.

[0023] In addition, the technical features involved in the different embodiments of the present application described below can be combined with each other as long as they do not conflict with each other.

[0024] The present application provides a foreign body recognition and impurity removal device based on image recognition technology. Figure 1 , including: a first frame 1, a material thinning mechanism, an image recognition system, and a foreign matter rejection mechanism.

[0025] The material thinning mechanism includes a second frame 2, a material leveling conveyor belt 3, a vibration motor 4, and a spring 5. A first driving wheel and a first driven wheel are respectively provided at both ends of the second frame 2. The first driving wheel is connected to the output shaft of the first motor. The material leveling conveyor belt 3 is sleeved on the first driving wheel and the first driven wheel at both ends of the second frame 2. The vibration motor 4 is arranged below the second frame 2 and fixedly connected to the second frame 2. The upper end of the spring 5 is connected to the second frame 2, and the lower end is connected to the first frame 1.

[0026] In some embodiments, the left end of the material conveyor belt 3 is higher than the right end.

[0027] In some embodiments, a plurality of vibration plates 6 are disposed on the second frame 2 , and the vibration plates 6 extend from one side of the second frame 2 to the other side along the width direction, and the upper and lower sides of the vibration plates 6 are in contact with the inner surface of the material leveling conveyor belt 3 .

[0028] Specifically, the material thinning mechanism consists of an inclined vibration plate 6 intersecting the material leveling conveyor belt 3, and its function is to loosen the thick material on the material leveling conveyor belt 3 through vibration and throw it at the left end onto the bidirectional foreign matter discharge conveyor belt 8 of the foreign matter removal mechanism to form a thin layer of material.

[0029] In some embodiments, there are at least four springs 5 ​​, which are respectively disposed at the left and right ends of both sides of the second frame 2 .

[0030] The foreign matter removal mechanism includes a third frame 7 and a bidirectional foreign matter discharge conveyor belt 8. The third frame 7 is connected to the first frame 1. The second driving wheel and the second driven wheel are respectively provided at both ends of the third frame 7. The second driving wheel is connected to the output shaft of the second motor. The bidirectional foreign matter discharge conveyor belt 8 is sleeved on the second driving wheel and the second driven wheel at both ends of the third frame 7. The right end of the bidirectional foreign matter discharge conveyor belt 8 is arranged below the left end of the material leveling conveyor belt 3.

[0031] In some embodiments, the foreign matter removal mechanism further includes: a foreign matter selection frame 9 , and the foreign matter selection frame 9 is arranged below the right end of the bidirectional foreign matter discharge conveyor belt 8 .

[0032] Specifically, the bidirectional foreign matter discharge conveyor belt 8 can be driven by the second driving wheel to transport to the left or right (i.e., forward or backward). Its function is that when foreign matter is detected, the bidirectional foreign matter discharge conveyor belt 8 transports to the right and discharges the abnormal material into the foreign matter picking vehicle below the right end of the bidirectional foreign matter discharge conveyor belt 8.

[0033] In some embodiments, the image recognition system includes a foreign object detection camera, which is connected to the first frame 1 and is arranged above the left end of the material leveling conveyor belt 3 and / or above the right end of the bidirectional foreign object discharge conveyor belt 8.

[0034] The function of the image recognition system is to analyze whether there are foreign objects in the material through image recognition algorithms and lock and mark them out.

[0035] In some embodiments, the image recognition system also includes: a recognition computer and a foreign object image display screen, wherein the recognition computer is connected to the foreign object detection camera and is used to obtain the detection image captured by the foreign object detection camera and display it on the foreign object image display screen.

[0036] In some embodiments, the foreign object detection camera includes a first camera 10 and a second camera 11, wherein the first camera 10 is disposed above the left end of the material leveling conveyor belt 3, and the second camera 11 is disposed above the right end of the bidirectional foreign object discharge conveyor belt 8.

[0037] In some embodiments, the foreign body identification and impurity removal device based on image recognition technology further includes: a conveyor belt programmable controller, the conveyor belt programmable controller is respectively connected to the identification computer, the first motor, and the second motor;

[0038] The conveyor belt programmable controller is used to receive instructions from the recognition computer and suspend the driving of the first motor to stop the material conveyor belt 3 from transporting materials after the first camera 10 detects that the foreign objects transported from right to left on the material conveyor belt 3 are thrown out of the material conveyor belt 3; when the second camera 11 detects that the foreign objects are thrown onto the bidirectional foreign object discharge conveyor belt 8, it receives instructions from the recognition computer to control the second motor to reverse and transport the bidirectional foreign object discharge conveyor belt 8 to the right end, thereby discharging the foreign objects into the foreign object selection box 9; when the second camera 11 detects that the foreign objects on the bidirectional foreign object discharge conveyor belt 8 are discharged out of the bidirectional foreign object discharge conveyor belt 8, it controls the second motor to rotate forward and the first motor to restart, so that the material continues to be transported from right to left.

[0039] In some embodiments, the foreign object detection camera is a JSA-8ELECPTZ3C series foreign object detection camera produced by JASIA.

[0040] In some embodiments, the algorithm of the image recognition technology is responsible for accurately identifying, locking and marking foreign objects in the material, providing accurate information for subsequent foreign object removal. The working principle and steps of this image recognition algorithm will be described in detail below.

[0041] First, the core of the algorithm lies in the processing and analysis of images. The first image data of the material is captured in real time by the first camera 10, and the first image data is preprocessed. In some embodiments, operations such as denoising and contrast enhancement are performed to improve the quality and clarity of the image, laying a foundation for subsequent recognition and analysis.

[0042] Next, the algorithm will use a specific feature extraction method to extract the first feature information of the material from the preprocessed first image data. In some embodiments, the first feature information includes the color, shape, texture, etc. of the material, which can reflect the basic properties and state of the material. By extracting these features, the algorithm can more accurately identify foreign matter in the material.

[0043] After the feature extraction is completed, the algorithm will use the classifier to classify and identify the extracted first feature information. In some embodiments, the classifier is usually built based on a machine learning or deep learning algorithm, which is learned and optimized through a large amount of training data and can accurately distinguish between materials and foreign objects. When new image data is input, the classifier will distinguish the materials and foreign objects in the first image data based on the previously learned knowledge and output the recognition result.

[0044] If the classifier detects a foreign object, the algorithm will immediately lock and mark the location of the foreign object. In some embodiments, this is achieved by drawing a rectangular box, circle or other mark on the image so that the operator can clearly see the location and size of the foreign object. In some embodiments, the algorithm will also record the relevant information of the foreign object (such as location, size, quantity, etc.) to provide accurate data support for the subsequent removal of foreign objects.

[0045] When it is detected that the locked and marked foreign objects are thrown out of the material leveling conveyor belt 3, a command is sent to the conveying programmable controller to suspend the driving of the first motor so that the material leveling conveyor belt 3 stops transporting materials.

[0046] The second image data of the material is captured in real time by the second camera 11, and the second image data are preprocessed, and the second characteristic information of the material is extracted from the preprocessed second image data. The extracted second characteristic information is classified and identified by a classifier, and the material and foreign matter in the second image data are distinguished, and the position of the foreign matter is locked and marked.

[0047] After the foreign object is locked and marked, an instruction is sent to the conveying programmable controller to control the second motor to reverse, so that the two-way foreign object discharge conveyor belt 8 is transmitted to the right end, thereby discharging the foreign object into the foreign object selection box 9; after it is detected that the foreign object is discharged from the two-way foreign object discharge conveyor belt 8, an instruction is sent to the conveying programmable controller to control the second motor to rotate forward and the first motor to restart, so that the material continues to be transported from right to left.

[0048] In summary, the image recognition algorithm plays a key role in the foreign body identification and impurity removal device based on image recognition technology. It realizes the accurate identification, locking and marking of foreign bodies in the material through image processing and analysis, providing accurate information and strong support for the subsequent foreign body removal.

[0049] In some embodiments, the construction and training methods of the machine learning or deep learning algorithms in the above schemes include:

[0050] In the above-mentioned foreign body identification and impurity removal device based on image recognition technology, the construction and training methods of machine learning or deep learning algorithms are crucial. The construction and training processes of these two algorithms will be introduced in detail below.

[0051] First, for the machine learning algorithm, its construction process includes two steps: feature extraction and classifier design. In the feature extraction stage, according to the characteristics of the material and foreign matter, an appropriate feature description method is selected, such as color, shape, texture and other features. Then, a feature vector is formed by calculating the eigenvalue of each pixel or area in the image. In some embodiments, in the classifier design stage, a classic machine learning algorithm such as support vector machine (SVM), decision tree and / or random forest is selected, and the classifier is trained using training data so that it can accurately distinguish between materials and foreign matter.

[0052] For deep learning algorithms, their construction mainly depends on the structural design of the neural network. Common deep learning network architectures include convolutional neural networks (CNNs) and recurrent neural networks (RNNs). During the construction process, it is necessary to select appropriate parameters such as the number of network layers, the number of neurons, and the activation function based on the specific recognition task and data characteristics. Then, the network is trained with a large amount of training data so that it can automatically learn the mapping relationship from the original image to the recognition result.

[0053] In terms of training methods, both machine learning and deep learning use supervised learning strategies. In some embodiments, specifically, a labeled training data set is first prepared, which contains a large number of material and foreign body images, as well as corresponding label information. Then, the training data set is input into the algorithm for training. During the training process, the algorithm will continuously adjust its own parameters and structure to minimize the error between the predicted results and the actual labels. Through multiple iterations and optimizations, the final machine learning or deep learning algorithm is obtained, and the algorithm can eventually learn the mapping relationship from image to recognition result, and achieve accurate recognition on new image data.

[0054] It should be noted that deep learning algorithms require a large amount of data and computing resources during the training process. In order to improve training efficiency and recognition accuracy, some optimization strategies are adopted, such as batch training, online learning, regularization, etc. In addition, the transfer learning method can also be used to migrate the model parameters that have been trained on other tasks to this task to accelerate the training process and improve performance.

[0055] In summary, the construction and training methods of machine learning or deep learning algorithms are a complex and delicate process that needs to be selected and adjusted according to specific tasks and data characteristics. Through reasonable algorithm design and training methods, accurate identification, locking and marking of foreign objects in materials can be achieved, providing strong support for subsequent foreign object removal.

[0056] In summary, compared with the prior art, the device and method provided in the embodiment of the present application fully consider the physical properties of foreign matter and the working conditions of the on-site equipment, and based on the image recognition technology, firstly, the thick material is fed into the material thinning mechanism and becomes a loose material thin layer through the vibration of the inclined surface, and the non-metallic small-volume foreign matter is identified by the image recognition system, and the abnormal material signal is transmitted to the recognition computer, so as to control the foreign matter removal mechanism to make the bidirectional foreign matter discharge conveyor belt 8 discharge the material wrapped with foreign matter, and the abnormal image is displayed on the foreign matter display screen at the same time. At this time, the on-site staff receives the abnormal material prompt, and searches and removes the non-metallic small-volume foreign matter in a targeted manner according to the abnormal material image. In addition, the on-site staff can use image recognition combined with the production process to analyze the cause of foreign matter, associate the cause with the location of the non-metallic small-volume foreign matter, and focus on the hidden dangers of the re-baking process. At this time, the device can accurately capture the non-metallic small-volume foreign matter in the high-speed moving material, make up for the shortcomings of the metal detector in the removal of non-metallic foreign matter, solve the problem of the difficulty of removing small-volume foreign matter by naked eye, reduce the cost of manual removal, and greatly improve the efficiency of impurity removal. At the same time, on-site staff can use image recognition technology and combine it with the production process to analyze the causes of foreign matter and the locations of related foreign matter, focus on checking potential danger areas, eliminate the generation of foreign matter in the production process from the source, and further improve production efficiency and product quality.

[0057] Note that, unless otherwise directly stated, all features disclosed in this specification (including any attached claims, abstracts and drawings) may be replaced by alternative features for achieving the same, equivalent or similar purposes. Therefore, unless otherwise explicitly stated, each feature disclosed is only an example of a group of equivalent or similar features. Where used, further, preferably, further and more preferably are simple beginnings for elaborating another embodiment based on the aforementioned embodiment, and the content of the further, preferably, further or more preferably followed by the combination with the aforementioned embodiment constitutes a complete construction of another embodiment. Several further, preferably, further or more preferably settings following the same embodiment can be arbitrarily combined to form another embodiment.

[0058] In the implementation of functions and steps, the corresponding functions and steps in various embodiments may also occur in a different order than shown. For example, two consecutive functions and steps may actually be performed or implemented substantially in parallel, and they may sometimes be performed or implemented in the opposite order, depending on the functions involved.

[0059] Although the present application has been described in detail above with general descriptions and specific embodiments, it is obvious to those skilled in the art that some modifications or improvements may be made to the present application. Therefore, these modifications or improvements made without departing from the spirit of the present application all fall within the scope of protection claimed in the present application.

Claims

1. A foreign body identification and impurity removal device based on image recognition technology, characterized in that: include: The first frame, material thinning mechanism, image recognition system, foreign body rejection mechanism, among which, The material thinning mechanism includes a second frame, a material leveling conveyor belt, a vibration motor, and a spring. A first driving wheel and a first driven wheel are respectively provided at two ends of the second frame. The first driving wheel is connected to the output shaft of the first motor. The material leveling conveyor belt is sleeved on the first driving wheel and the first driven wheel at two ends of the second frame. The vibration motor is arranged below the second frame and is fixedly connected to the second frame. The upper end of the spring is connected to the second frame, and the lower end is connected to the first frame. The foreign body rejection mechanism includes a third frame and a bidirectional foreign body discharge conveyor belt, the third frame is connected to the first frame, and two ends of the third frame are respectively provided with a second driving wheel and a second driven wheel, the second driving wheel is connected to the output shaft of the second motor, and the bidirectional foreign body discharge conveyor belt is sleeved on the second driving wheel and the second driven wheel at two ends of the third frame, and the right end of the bidirectional foreign body discharge conveyor belt is arranged below the left end of the material leveling conveyor belt; The image recognition system comprises a foreign object detection camera, which is connected to the first frame and arranged above the left end of the material leveling conveyor belt and / or above the right end of the bidirectional foreign object discharge conveyor belt.

2. The foreign matter identification and impurity removal device based on image recognition technology according to claim 1 is characterized in that: The left end of the material leveling conveyor belt is higher than the right end.

3. The foreign matter identification and impurity removal device based on image recognition technology according to claim 1 is characterized in that: The second frame is provided with a plurality of vibration plates, which extend from one side of the second frame to the other side along the width direction, and the upper and lower sides of the vibration plates are in contact with the inner surface of the material leveling conveyor belt.

4. The foreign matter identification and impurity removal device based on image recognition technology according to claim 1 is characterized in that: There are at least four springs, which are respectively arranged at the left and right ends of both sides of the second frame.

5. The foreign matter identification and impurity removal device based on image recognition technology according to claim 1 is characterized in that: The foreign body removal mechanism also includes: A foreign matter selection frame is arranged below the right end of the bidirectional foreign matter discharge conveyor belt.

6. The foreign matter identification and impurity removal device based on image recognition technology according to claim 1 is characterized in that: The image recognition system further comprises: An identification computer and a foreign body image display screen are connected to the foreign body detection camera to obtain the detection image collected by the foreign body detection camera and display it on the foreign body image display screen.

7. The foreign matter identification and impurity removal device based on image recognition technology according to claim 1 is characterized in that: The foreign object detection camera includes a first camera and a second camera, wherein the first camera is arranged above the left end of the material leveling conveyor belt, and the second camera is arranged above the right end of the bidirectional foreign object discharge conveyor belt.

8. The foreign matter identification and impurity removal device based on image recognition technology according to claim 6 is characterized in that: Also includes: A conveyor belt programmable controller is connected to the identification computer, the first motor and the second motor respectively.

9. The foreign matter identification and impurity removal device based on image recognition technology according to claim 1 is characterized in that: The foreign object detection camera is a JSA-8ELECPTZ3C series foreign object detection camera from JASIA.

Images