An image recognition-based roller failure detection method

Abstract

The present application belongs to the technical field of safety monitoring of ginning machines, and specifically discloses a ginning machine fault detection method based on image recognition, which comprises: acquiring initial cotton roll images of the opening area of the cotton roll cavity in real time by using a visual camera, and denoising the initial cotton roll image information to obtain cotton roll outer surface images; obtaining multiple groups of cotton roll flow rate values according to the cotton roll outer surface images; judging whether the ginning machine is in a cotton carding fault state according to the multiple groups of cotton roll flow rate values; if so, using a preset convolutional neural network model to identify the cotton carding degree, determining the cotton carding degree grade, determining the corresponding fault warning operation from the preset warning strategy, and warning according to the fault warning operation. Through double fault analysis of the cotton roll flow rate and the cotton roll image, the cotton carding state is determined, and different fault warning operations are correspondingly set. Full-automatic identification and control are realized, the accuracy and timeliness of cotton carding identification are improved, the cotton ginning efficiency is improved, and the operation safety is ensured.

Description

Technical Field

[0001] This invention relates to the field of cotton gin safety monitoring technology, specifically to a cotton gin fault detection method based on image recognition. Background Technology

[0002] A cotton gin is a type of cotton processing machinery used to separate lint from seed cotton. Commonly used gins include saw gins and roller gins. The saw gin utilizes a high-speed rotating (approximately 12-13 m / s) circular saw blade to hook and separate cotton fibers from the seed cotton through the gaps between the ribs. There are two types: brush-type and air-flow-type. Its main working components include feeding rollers, a cleaning mechanism, a ginning chamber, ginning ribs, a saw blade cylinder, a brush roller or air-flow suction nozzle, and a cotton collection box. Seed cotton is fed into the cleaning mechanism via paired feeding rollers, then enters the front chamber of the gin. It is thrown against the saw blade by the cotton-distributing rollers, while large debris such as bolls fall through. The seed cotton is hooked by the saw teeth and carried into the ginning chamber. The saw blade rotates rapidly, causing the mutually pulling seed cotton to form a rotating seed cotton lap. The saw teeth hook the fibers and rotate them. After passing through the gap between adjacent ginning ribs, the cotton fibers are brushed off by the brush roller (or airflow) and sent into the cotton collection box. The cotton seeds are blocked by the ribs and move down along the rib surface between the two saw blades. They are then discharged from the machine through the cotton seed comb.

[0003] During the operation of the cotton gin, cotton jamming is prone to occur at the junction of the saw blade cylinder and the ginning ribs. Once cotton jamming occurs, the cotton will heat up and catch fire in a very short time, causing equipment failure. Currently, cotton jamming is identified by manual observation. However, due to the limitations of human vision and the long reaction time of the human body, it is impossible to identify the problem in time and take relevant preventive measures. This leads to frequent occurrences of cotton jamming and fire in the cotton gin, reducing cotton ginning efficiency and posing safety hazards. Summary of the Invention

[0004] The purpose of this invention is to provide a fault detection method for cotton ginning machines based on image recognition, so as to solve the problem of cotton jamming in cotton ginning machines not being identified in a timely manner.

[0005] To achieve the above objectives, the basic solution provided by this invention is: a fault detection method for a cotton ginning machine based on image recognition, comprising:

[0006] S1: Use a vision camera to acquire the initial cotton roll image of the opening area of ​​the cotton roll cavity in real time, and reduce the noise of the initial cotton roll image information to obtain the outer surface image of the cotton roll.

[0007] S2. Based on the image of the outer surface of the cotton roll, use a vision camera to obtain multiple sets of cotton roll flow rate values ​​within a preset time period;

[0008] S3: Based on multiple sets of cotton roll flow rate values, determine whether the cotton gin is in a cotton jamming fault state. If it is in a cotton jamming fault state, based on the cotton roll outer surface image, use a preset convolutional neural network model to identify the degree of fault and determine the current cotton jamming level of the cotton gin. Among them, the cotton jamming level includes level one cotton jamming fault and level two cotton jamming fault.

[0009] S4: Based on the level of cotton jamming, determine the corresponding fault warning operation from the preset warning strategy;

[0010] S5: Issue a warning based on the fault warning operation.

[0011] Because cotton jamming easily occurs at the junction of the serrated cylinder and the ginning ribs, the cotton roll rotation speed slows down, and the outer surface of the cotton roll bulges or collapses. Since the junction of the serrated cylinder and the ginning ribs is inside the cotton roll cavity and cannot be directly detected, this invention utilizes a vision camera to monitor the cotton roll image in the opening area of ​​the cotton roll cavity in real time. The vision camera's internal functional coding obtains multiple sets of cotton roll flow rate values ​​within a preset time period. Based on these multiple sets of flow rate values, it determines whether the ginning machine is in a cotton jamming fault state. If a cotton jamming fault state is detected, a preset convolutional neural network model is used to identify the degree of fault, determining the current cotton jamming level of the ginning machine. The cotton jamming level and fault warning operation are interlocked. Different fault states correspond to different fault warning operations, improving the practicality of fault identification. The fully automated identification and control solves the problem of untimely identification of cotton jamming in the ginning machine, improves ginning efficiency, and ensures the safe operation of the ginning machine.

[0012] Preferably, step S1 involves denoising the initial cotton roll image, specifically including:

[0013] Perform wavelet transform on the initial cotton roll image to obtain the high-frequency and low-frequency components of the initial cotton roll image;

[0014] The portion of the high-frequency component with wavelet coefficients below a preset wavelet coefficient threshold is set to zero to obtain the noise-reduced high-frequency component.

[0015] Wavelet reconstruction is performed on the low-frequency component and the noise-reduced high-frequency component to obtain the image of the outer surface of the cotton roll.

[0016] Understandably, useful signals manifest as low-frequency or stable signals, while noise manifests as high-frequency signals. By using wavelet transform, the cotton image removes unnecessary interference information in the high-frequency components, thus denoising the initial cotton roll image and improving its clarity and reliability.

[0017] Preferably, the determination of the cotton gin malfunction in step S3 specifically includes:

[0018] Arrange multiple sets of cotton roll flow rate values ​​in chronological order;

[0019] The multiple sets of cotton roll flow rate values ​​are compared with the preset fault speed range to determine whether the multiple sets of cotton roll flow rate values ​​fall continuously within the preset fault speed range.

[0020] If the speed does not continuously fall within the preset fault speed range, it is determined that the current cotton gin is not stuck; if it continuously falls within the preset fault speed range, it is determined that the current cotton gin is in a cotton stuck fault state.

[0021] By setting a preset fault speed range, the obtained multiple sets of cotton roll flow rate values ​​are compared with the preset fault speed range. If they are continuously within the range, it indicates that there may be cotton jamming at the junction of the sawtooth cylinder and the ginning rib. If they continuously fall within the preset fault speed range, it is determined that the current ginning machine is in a cotton jamming fault state. This realizes the identification of the degree of cotton jamming and provides a data basis for subsequent fault warning operations interlocked with the degree of cotton jamming.

[0022] Preferably, the pre-training of the convolutional neural network model in step S3 specifically includes:

[0023] Acquire historical images of cotton jams in the opening area of ​​the cotton roll cavity;

[0024] Historical cotton jam images are classified into primary cotton jam images and secondary cotton jam images;

[0025] Based on the primary and secondary cotton jam images, establish training and validation sets;

[0026] The convolutional neural network model is trained using the training set to obtain the trained convolutional neural network model;

[0027] The validation set is used to check whether the trained convolutional neural network model is qualified. If it is qualified, the convolutional neural network model is output; if it is not qualified, the convolutional neural network model is retrained using the training set until the convolutional neural network model is qualified.

[0028] Based on historical cotton jam images, the image information is divided into Level 1 and Level 2 cotton jam images. Level 1 cotton jam images correspond to a low degree of cotton jamming, so they are set as Level 1. Level 2 cotton jam images correspond to a high degree of cotton jamming, so they are set as Level 2. Based on these two types of images, training and validation sets for the model are established. The initial convolutional neural network model is trained using the training set, and the validation set is used to verify whether the trained convolutional neural network model is qualified, so as to obtain a model that can realize image fault recognition and improve the accuracy of model fault recognition.

[0029] Preferably, the preset convolutional neural network model includes an input layer, a convolutional layer, a pooling layer, a fully connected layer, and an output layer;

[0030] Step S3, which uses a pre-defined convolutional neural network model to identify the degree of fault, specifically includes the following steps:

[0031] The image of the outer surface of the cotton roll is normalized to obtain standard image information; the standard image information includes pixel information and color channel information.

[0032] Standard image information will be input into the input layer;

[0033] The convolutional layer receives standard image information output from the input layer and uses a convolutional kernel of a preset size and the ReLU function as its activation function to extract features from the standard image information to obtain the initial image features;

[0034] The pooling layer receives the initial image features output by the convolutional layer and performs average pooling on the initial image features to obtain the target image features;

[0035] The fully connected layer receives the target image features output by the pooling layer, flattens the target image features into a one-dimensional vector, uses the Softmax function to classify them, obtains the probability corresponding to each cotton jamming level category, filters out the cotton jamming level category with the highest probability, and uses the cotton jamming level category with the highest probability as the cotton jamming level corresponding to the current cotton ginning machine image.

[0036] The output layer receives the cotton jamming level from the fully connected layer and outputs the cotton jamming level.

[0037] The convolutional neural network (CNN) model is configured with an input layer, convolutional layer, pooling layer, fully connected layer, and output layer. Since the CNN model uses gradient descent for learning, the input features need to be standardized. Therefore, this invention normalizes the cotton roll outer surface image before it is input to the input layer to meet the data processing requirements of the CNN model. The standardized image information is then input to the input layer, which transmits the data to the convolutional layer. A pre-sized convolutional kernel and ReLU function are used as activation functions for image feature extraction. The feature is then averaged through the pooling layer to obtain the target image features, which are then input to the fully connected layer to flatten the target image features into a one-dimensional vector. The Softmax function is used for classification to obtain the cotton jamming level corresponding to the current ginning machine image. Finally, the output layer outputs the cotton jamming level, enabling cotton jamming fault analysis based on the cotton roll outer surface image. This avoids the limitations and lag of current manual identification, improving the efficiency and accuracy of cotton jamming fault identification.

[0038] Preferably, the fault warning operation corresponding to the first-level cotton jam fault in the preset early warning strategy specifically includes:

[0039] The system generates and sends a first-level traffic light flashing command to the traffic light, and generates a corresponding first-level voice broadcast content. The broadcast content is then broadcast using a broadcasting device to achieve first-level traffic light warning and first-level voice broadcast warning.

[0040] In the event of a level 1 cotton jam, the cotton ginning speed may slow down, and slight bulges or dents may appear on the outer surface of the cotton roll. However, the cotton gin can still operate normally, and the degree of failure is low. Therefore, this invention only provides voice broadcast warnings and signal light warnings to remind operators that the cotton gin may have experienced a slight cotton jam.

[0041] Preferably, the fault warning operation corresponding to the secondary cotton jamming fault in the preset early warning strategy specifically includes:

[0042] Generate and send the cotton ginning machine unpacking command to the cotton ginning machine, so that the cotton ginning machine can be unpacked;

[0043] The system generates and sends a secondary signal light flashing command to the signal light, and generates corresponding secondary voice broadcast content. The system then broadcasts the secondary voice broadcast content using a broadcasting device to achieve secondary signal light prompts and secondary voice broadcast prompts.

[0044] In the event of a secondary cotton jamming fault, a large cotton clump may be stuck between the sawtooth cylinder and the ginning ribs, resulting in a significant reduction in cotton ginning speed and the appearance of large bulges or dents on the outer surface of the cotton roll. This invention directly controls the opening of the ginning machine and provides signal lights and voice prompts to alert the operator to handle the cotton jamming problem. Attached Figure Description

[0045] Figure 1 This is a diagram of the internal structure of a cotton ginning machine. Detailed Implementation

[0046] The present invention will be further described in detail below through specific embodiments:

[0047] The reference numerals in the accompanying drawings of the instruction manual include: 1. ginning machine; 2. toothed drum; 3. ginning ribs; 4. vision camera; 5. cotton roll cavity.

[0048] like Figure 1 As shown in the figure, this embodiment provides a fault detection method for cotton ginning machines based on image recognition, which specifically includes the following:

[0049] S1: Use vision camera 4 to acquire the initial cotton roll image of the opening area of ​​cotton roll cavity 5 in real time, and reduce the noise of the initial cotton roll image information to obtain the outer surface image of the cotton roll.

[0050] Because cotton is prone to jamming at the junction of the sawtooth cylinder 2 and the ginning ribs 3, the speed of cotton rotation will slow down when cotton is jammed. Also, because the junction of the sawtooth cylinder 2 and the ginning ribs 3 is inside the cotton roll cavity 5, it cannot be directly detected. In this embodiment, a vision camera 4 is mounted on the outer wall of the ginning machine 1. The camera is aimed at the opening area of ​​the cotton roll cavity 5 to take pictures, and the vision camera 4 captures images through the camera.

[0051] The noise reduction of the initial cotton roll image information specifically includes: performing wavelet transform on the initial cotton roll image to obtain the high-frequency and low-frequency components of the initial cotton roll image; setting the wavelet coefficients of the high-frequency component below the preset wavelet coefficient threshold to zero to obtain the noise-reduced high-frequency component; and performing wavelet reconstruction on the low-frequency component and the noise-reduced high-frequency component to obtain the outer surface image of the cotton roll. This removes unnecessary interference information in the high-frequency component, thereby achieving noise reduction of the outer surface image of the cotton roll and improving the image clarity and reliability.

[0052] S2. Based on the image of the outer surface of the cotton roll, use a vision camera to obtain multiple sets of cotton roll flow rate values ​​within a preset time period.

[0053] After the vision camera 4 acquires an image of the outer surface of the cotton roll, it calibrates and establishes a mapping relationship between the image pixel coordinates and the world coordinates of the corresponding points in space, extracts and selects image features, obtains the actual displacement of the image features over a period of time, and then calculates the cotton roll flow rate value.

[0054] S2: Based on multiple sets of cotton roll flow rate values, determine whether the cotton gin 1 is in a cotton jamming fault state. If it is in a cotton jamming fault state, based on the cotton roll outer surface image, use a preset convolutional neural network model to identify the degree of fault and determine the current cotton jamming level of the cotton gin 1. Among them, the cotton jamming level includes level one cotton jamming fault and level two cotton jamming fault.

[0055] This embodiment classifies cotton jamming into Level 1 and Level 2 faults. In Level 1, the flow rate on the outer surface of the cotton roll may slow down, and slight bulges or dents may appear on the outer surface. However, the ginning machine 1 can still operate normally; the fault level is low, and the machine can continue operating without opening the box. In Level 2, a large cotton ball may be stuck between the toothed cylinder 2 and the ginning ribs 3, causing a significant reduction in the flow rate on the outer surface of the cotton roll, resulting in larger bulges or dents. This fault level is high, requiring the ginning machine 1 to be opened to address the cotton jam. It should be noted that in this embodiment, the ginning machine 1 is kept running continuously when the box is opened.

[0056] The identification of cotton jam fault status specifically includes: arranging multiple sets of cotton roll flow rate values ​​in chronological order; comparing the acquired multiple sets of cotton roll flow rate values ​​with the preset fault speed range by setting a preset fault speed range; if the value is within the range, it indicates that there may be cotton jamming at the junction of the sawtooth cylinder 2 and the ginning rib 3; if the value continuously falls within the preset fault speed range, it is determined that the current ginning machine is in a cotton jam fault state, thus realizing the identification of the degree of cotton jamming and providing a data basis for subsequent fault warning operations interlocked with the degree of cotton jamming.

[0057] The pre-training of the convolutional neural network model specifically includes: acquiring historical images of cotton jamming in the opening area of ​​the cotton roll cavity; classifying these images into primary and secondary jamming images; primary jamming faults are characterized by small bulges or indentations on the outer surface of the cotton roll, indicating low jamming severity, and are therefore classified as primary jamming faults; secondary jamming faults are characterized by large bulges or indentations on the outer surface of the cotton roll, indicating high jamming severity, and are therefore classified as secondary jamming faults; training and validation sets are then established based on the primary and secondary jamming images; the convolutional neural network model is trained using the training set to obtain the trained convolutional neural network model; and the validation set is used to verify whether the trained convolutional neural network model is qualified. If qualified, the convolutional neural network model is output; if not qualified, the convolutional neural network model is retrained using the training set until it is qualified, thus obtaining a model capable of image fault recognition and improving the accuracy of fault recognition.

[0058] In this embodiment, the convolutional neural network model is configured with the following layers in sequence: input layer, convolutional layer, pooling layer, fully connected layer, and output layer. Since the convolutional neural network model uses the gradient descent algorithm for learning, the input features need to be standardized. Therefore, in this embodiment, the outer surface image of the cotton roll is normalized before being input to the input layer to meet the data processing requirements of the convolutional neural network model. The normalized standard image information includes pixel information and color channel information.

[0059] The specific process of convolutional neural network fault identification includes: inputting normalized standard image information into the input layer; the input layer transmits the data to the convolutional layer, which uses a pre-sized convolutional kernel and the ReLU function as its activation function to extract image features; then, the features are processed by the pooling layer to obtain the target image features; the target image features are then input into the fully connected layer to flatten them into a one-dimensional vector; the Softmax function is used for classification to obtain the probability corresponding to each cotton jamming level; the cotton jamming level corresponding to the highest probability is selected; the cotton jamming level corresponding to the highest probability is used as the cotton jamming level corresponding to the current cotton gin image; and the output layer outputs the cotton jamming level to achieve cotton jamming level analysis based on the outer surface image of the cotton roll, avoiding the limitations and lag of current manual identification, and improving the efficiency and accuracy of cotton jamming fault identification. For example, in this embodiment, the layers of the convolutional neural network model are set as input layer, convolutional layer, pooling layer, convolutional layer, pooling layer... convolutional layer, pooling layer, fully connected layer, and output layer. A standard 32×32 image is input to the input layer. The convolutional layers use 5×5 convolutional kernels to convolve the image from the input layer, and then pass it through the ReLU activation function to obtain 20 two-dimensional feature maps of size 28×28. The pooling layers perform average pooling on all 2×2 sub-blocks in the convolutional layers to obtain 20 two-dimensional feature maps of size 14×14. The small two-dimensional feature maps are processed, and the subsequent convolutional and pooling layers repeat the steps described above. This results in 50 two-dimensional feature maps of size 5×5. A fully connected layer flattens these 50 5×5 feature maps into a one-dimensional vector. The Softmax function is then used for classification to obtain the probability corresponding to each cotton ginning severity category. The category with the highest probability is selected as the cotton ginning severity level for the current image. The output layer then outputs the cotton ginning severity level. The activation functions ReLU and Softmax are shown in Equations 1 and 2, respectively.

[0060] f(x)=max(0,x) (1)

[0061] Here, x is the input, and the output is the larger value between x and 0, which effectively solves the problem of gradient vanishing.

[0062]

[0063] Among them, z j Let represent the score of the j-th cotton level category, softmax(z). j This represents the estimated probability of the j-th cotton blemish severity category, where K is the dimension of the vector output by the softmax function, the sum of the vector elements is 1, and the final output is the probability value corresponding to each cotton blemish severity level.

[0064] S3: Based on the level of cotton jamming, determine the corresponding fault warning operation from the preset warning strategy.

[0065] The pre-set early warning strategy for the first-level cotton jam fault is as follows: generate and send a first-level signal light flashing command to the signal light, and generate the corresponding first-level voice broadcast content. The broadcast content is then broadcast using a broadcasting device to realize the first-level signal light warning and the first-level voice broadcast warning, so as to remind the operator that the cotton gin 1 may have a slight cotton jam.

[0066] The pre-set early warning strategy for the secondary cotton jam fault is as follows: generate and send an opening command to ginning machine 1 to ginning machine 1, so that ginning machine 1 opens the box, then generate and send a secondary signal light flashing command to the signal light, and generate the corresponding secondary voice broadcast content. According to the secondary voice broadcast content, broadcast it using the broadcasting equipment to realize the secondary signal light prompt and the secondary voice broadcast prompt, prompting the operator to deal with the cotton jam problem.

[0067] S5: Issue a warning based on the fault warning operation.

[0068] The above descriptions are merely embodiments of the present invention, and common knowledge regarding specific structures and characteristics is not elaborated upon here. It should be noted that those skilled in the art can make various modifications and improvements without departing from the structure of the present invention, and these should also be considered within the scope of protection of the present invention. These modifications and improvements will not affect the effectiveness of the present invention or the practicality of the patent. The scope of protection claimed in this application should be determined by the content of its claims, and the specific embodiments described in the specification can be used to interpret the content of the claims.

Claims

1. A fault detection method for a cotton ginning machine based on image recognition, characterized in that, include: S1: Use a vision camera (4) to acquire an initial cotton roll image of the opening area of ​​the cotton roll cavity (5) in real time, and reduce the noise of the initial cotton roll image information to obtain an image of the outer surface of the cotton roll; S2. Based on the image of the outer surface of the cotton roll, the visual camera (4) is used to obtain multiple sets of cotton roll flow rate values ​​within a preset time period. S3: Based on the multiple sets of cotton roll flow rate values, determine whether the cotton gin (1) is in a cotton jamming fault state. If it is in a cotton jamming fault state, then based on the outer surface image of the cotton roll, use the preset convolutional neural network model to identify the degree of fault and determine the current cotton jamming level of the cotton gin (1); wherein, the cotton jamming level includes first-level cotton jamming fault and second-level cotton jamming fault. S4: Based on the level of cotton jamming, determine the corresponding fault warning operation from the preset warning strategy; S5: Issue a warning based on the aforementioned fault warning operation; The specific steps for determining the malfunction of the cotton gin in step S3 include: Arrange the multiple sets of cotton roll flow rate values ​​in chronological order; The multiple sets of cotton roll flow rate values ​​are compared with the preset fault speed range to determine whether the multiple sets of cotton roll flow rate values ​​fall continuously within the preset fault speed range. If the speed does not fall continuously within the preset fault speed range, it is determined that the current cotton gin (1) is not stuck; if it falls continuously within the preset fault speed range, it is determined that the current cotton gin (1) is in a cotton stuck fault state.

2. The image recognition-based fault detection method for a cotton ginning machine according to claim 1, characterized in that, Step S1 involves denoising the initial cotton roll image, specifically including: The initial cotton roll image is subjected to wavelet transform to obtain the high-frequency and low-frequency components of the initial cotton roll image; The portion of the high-frequency component with wavelet coefficients lower than a preset wavelet coefficient threshold is set to zero to obtain the noise-reduced high-frequency component. The low-frequency component and the noise-reduced high-frequency component are reconstructed using wavelet reconstruction to obtain an image of the outer surface of the cotton roll.

3. The image recognition-based fault detection method for a cotton ginning machine according to claim 1, characterized in that, The pre-training of the convolutional neural network model in step S3 specifically includes: Acquire historical images of cotton jams in the opening area of ​​the cotton roll cavity (5); Historical cotton jam images are classified into primary cotton jam images and secondary cotton jam images; Based on the primary cotton jamming image and the secondary cotton jamming image, a training set and a validation set are established; The convolutional neural network model is trained using the training set to obtain the trained convolutional neural network model; The training convolutional neural network model is tested using the validation set to determine if it is qualified. If it is qualified, the convolutional neural network model is output. If it is not qualified, the convolutional neural network model is retrained using the training set until it is qualified.

4. The image recognition-based fault detection method for a cotton ginning machine according to claim 3, characterized in that, The preset convolutional neural network model includes an input layer, a convolutional layer, a pooling layer, a fully connected layer, and an output layer. Step S3, which uses a pre-defined convolutional neural network model to identify the degree of fault, specifically includes the following steps: The outer surface image of the cotton roll is normalized to obtain standard image information; wherein, the standard image information includes pixel information and color channel information; The standard image information is input into the input layer; The convolutional layer receives the standard image information output by the input layer, and uses a convolutional kernel of a preset size and the ReLU function as its activation function to extract features from the standard image information to obtain the initial image features; The pooling layer receives the initial image features output by the convolutional layer and performs average pooling on the initial image features to obtain the target image features; The fully connected layer receives the target image features output by the pooling layer, flattens the target image features into a one-dimensional vector, classifies them using the Softmax function, obtains the probability corresponding to each cotton jamming level category, filters out the cotton jamming level category corresponding to the highest probability, and uses the cotton jamming level category corresponding to the highest probability as the cotton jamming level corresponding to the current cotton ginning machine image. The output layer receives the cotton jamming level output by the fully connected layer and outputs the cotton jamming level.

5. The image recognition-based fault detection method for a cotton ginning machine according to claim 1, characterized in that, The specific fault warning operations corresponding to the first-level cotton jamming fault in the preset early warning strategy include: A primary signal light flashing command is generated and sent to the signal light, and a corresponding primary voice broadcast content is generated. The broadcast content is then broadcast using a broadcasting device to achieve primary signal light warning and primary voice broadcast warning.

6. The image recognition-based fault detection method for a cotton ginning machine according to claim 1, characterized in that... The specific fault warning operations corresponding to the secondary cotton jamming fault in the preset early warning strategy include: Generate and send the cotton ginning machine unpacking command to the cotton ginning machine (1) so that the cotton ginning machine (1) is unpacked; A secondary signal light flashing command is generated and sent to the signal light, and a corresponding secondary voice broadcast content is generated. The broadcast content is then broadcast using a broadcasting device to achieve secondary signal light prompts and secondary voice broadcast prompts.

Images