CN110866561A - Plastic bottle color sorting method based on image recognition - Google Patents
Plastic bottle color sorting method based on image recognition Download PDFInfo
- Publication number
- CN110866561A CN110866561A CN201911130056.8A CN201911130056A CN110866561A CN 110866561 A CN110866561 A CN 110866561A CN 201911130056 A CN201911130056 A CN 201911130056A CN 110866561 A CN110866561 A CN 110866561A
- Authority
- CN
- China
- Prior art keywords
- data set
- image
- image data
- layer
- network model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a plastic bottle color sorting method based on image recognition, which comprises the following steps: and S1, collecting the plastic bottle image, and dividing the image into a training image data set, a verification image data set and an image data set to be detected according to the proportion after carrying out denoising, averaging, defogging and cutting operations on the image. S2, establishing a deep learning network model for plastic bottle color sorting, inputting the training image data set into the network model for training, and determining the deep learning network model and model parameters. And S3, inputting the verification image data set into the determined deep learning network model, and checking the deep learning network model. And S4, inputting the image data set to be detected into the deep learning network model to obtain the plastic bottle color classification result in the image data set to be detected. The invention utilizes the deep learning image detection and classification to carry out real-time sorting on the colors of the plastic bottles, and can realize the plastic bottle color sorting result with high efficiency and strong accuracy.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a plastic bottle color sorting method based on image recognition.
Background
With the development of times and the improvement of living standard of people, the production and consumption of the beverage are continuously increased. For the treatment of plastic bottles, most of the existing methods are burying and burning, because the natural degradation time of the plastic bottles is extremely long, and harmful substances are generated in the degradation process; the combustion method can generate a large amount of harmful gases and the like to cause serious pollution to the environment, harm the health of people and cause resource waste. Therefore, the recycling of plastic bottles is carried out in China, and the plastic bottles are classified by colors in order to ensure the quality of recycled materials. At present, the color classification of plastic bottles in China mainly depends on manual sorting, and the efficiency is not high, so that the color classification is a work which consumes labor, time and cost. The deep learning method is utilized to greatly improve the working efficiency of detecting and classifying the colors of the plastic bottles, and simultaneously save a large amount of labor and cost.
The current color identification methods include: performing operations such as segmentation on the image through edge detection and image binarization processing, and determining color classification by using the color difference relation between the colors of other segmented regions and the colors of the reference region; and performing HSV color space conversion on the RGB color image through weight blocking, and performing color classification according to the H and V values. The existing method is not suitable for experiments with large data sets, and the classification speed and accuracy are not high.
Disclosure of Invention
The invention provides the plastic bottle color sorting method which is high in accuracy, good in rapidity and strong in practicability and is suitable for the color change of the plastic bottles in different environments.
A plastic bottle color sorting method based on image recognition comprises the following steps:
s1, collecting a plastic bottle image, and after carrying out denoising, averaging, defogging and cutting operations on the image, dividing the image into a training image data set, a verification image data set and an image data set to be detected according to a proportion;
s2, establishing a deep learning network model for plastic bottle color sorting, inputting a training image data set into the network model for training, and determining the deep learning network model and model parameters;
s3, inputting the verification image data set into the determined deep learning network model, and checking the deep learning network model;
and S4, inputting the image data set to be detected into the deep learning network model to obtain the plastic bottle color classification result in the image data set to be detected.
Further, the step S1 includes the following steps:
s11, carrying out image acquisition on plastic bottles with different colors by using an industrial camera to form a first data set;
s12, carrying out Gaussian filtering denoising on the image in the first data set;
s13, carrying out gray level averaging and defogging operation treatment on the image processed in the S12;
s14, cutting the graph processed in the S13 according to the size of 224 × 224, storing the cut image into a first data set, replacing the original image and forming a second data set;
s15, classifying the plastic bottles according to red, green, blue, purple, yellow, black, white and transparent colors for the image of the second data set in the S14, labeling, rotating in different directions, mirroring, increasing contrast and the like, and then, carrying out the following operations on the second data set according to the ratio of 8: 1: 1, and obtaining a training image data set, a verification image data set and an image data set to be detected after preprocessing.
Further, the step S2 includes the following steps:
s21, determining a convolutional layer and a pooling layer of the input image;
s22, extracting the characteristics of the color of the plastic bottle through a residual error network;
s23, inputting the extracted feature map into a global average pooling layer, and inputting the pooling features into a full-connection layer for plastic bottle color classification;
s24, setting parameters in the training algorithm, including learning rate, training times and iteration times;
and S25, inputting the training image data set into the established network model, and determining the deep learning network model and the model parameters.
Further, the convolutional layers in S21 include a first convolutional layer and a second convolutional layer, where the size of the convolutional core in the first convolutional layer is 3 × 3, and the step size is 2; the convolution kernel in the second convolution layer has a size of 3 x 3 and a step size of 1.
Further, the residual error network in S22 is provided with four residual error blocks, and the residual error blocks are provided with a third convolutional layer, a fourth convolutional layer and a fifth convolutional layer.
Further, the output calculation formula of the residual block is as follows:
Y＝relu3([relu2(relu1(X,W1),W2)]*W3+X*Wt)
wherein X is the input, Y is the output, W1、W2、W3For each convolutional layer, relu1For the excitation function of the third layer convolutional layer output, relu2For the excitation function output by the convolutional layer of the fourth layer, relu3Wt is the linear transformation matrix for the excitation function of the residual block output.
Further, the convolution kernel of the third convolution layer is 1 × 1 with a step size of 2, the convolution kernel of the fourth convolution layer is 3 × 3 with a step size of 1, and the convolution kernel of the fifth convolution layer is 1 × 1 with a step size of 1.
Further, the step S25 includes the following steps:
s251, judging whether the set training times is reached, if so, obtaining a result, otherwise, repeating the S251;
s252, judging whether the set iteration times are reached, if so, finishing the training, otherwise, repeating S251;
and S253, determining a deep learning network model and model parameters.
The invention has the beneficial effects that:
1. the image is subjected to Gaussian filtering denoising, gray level equalization and defogging treatment, so that the problem of interference on the image under different light problems is solved, and the quality of a data set is improved;
2. in the image input, two 3 × 3 convolution kernels (the step length of the first convolution kernel is 2, and the step length of the second convolution kernel is 1) are used for replacing a 7 × 7 convolution kernel with the step length of 2, the number of layers of the network is increased, and an excitation function is added between the layers, so that the nonlinear expression capability of the network is enhanced, and the number of parameters is reduced;
3. for the residual block, the calculation amount is also reduced by using the method of reducing the network dimension before restoring the network dimension.
Drawings
FIG. 1 is an overall flow chart of the present invention;
FIG. 2 is a detailed flowchart of the present invention S1;
FIG. 3 is a detailed flowchart of the present invention S2;
fig. 4 is a diagram of a deep learning network model structure established by the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be described clearly and completely with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments obtained by a person skilled in the art without making any inventive step are within the scope of the present invention.
Referring to fig. 1 to 4, a plastic bottle color sorting method based on image recognition includes the following steps:
and S1, collecting the plastic bottle image, and dividing the image into a training image data set, a verification image data set and an image data set to be detected according to the proportion after carrying out denoising, averaging, defogging and cutting operations on the image. The method comprises the following specific steps:
s11, carrying out image acquisition on plastic bottles with different colors by using an industrial camera to form a first data set;
s12, carrying out Gaussian filtering denoising on the image in the first data set;
and S13, carrying out gray level averaging operation on the image processed in the S12, wherein the specific method comprises the following steps: dividing the color image into three independent single channels, then carrying out gray histogram equalization on each single channel, and merging the equalized images into one channel; then, defogging operation is carried out on the image to restore the image into a color image;
s14, clipping the graphics processed in S13 according to the size of 224 × 224, storing the clipped images into a first data set, replacing the original images, and forming a second data set;
s15, classifying the plastic bottles according to red, green, blue, purple, yellow, black, white and transparent colors for the image of the second data set in the S14, labeling, rotating in different directions, mirroring, increasing contrast and the like, and then, carrying out the following operations on the second data set according to the ratio of 8: 1: 1, and obtaining a training image data set, a verification image data set and an image data set to be detected after preprocessing.
S2, establishing a deep learning network model for plastic bottle color sorting, inputting the training image data set into the network model for training, and determining the deep learning network model and model parameters. The method comprises the following specific steps:
s21, determining the convolution layer and the pooling layer of the input image. In order to enhance the nonlinear expression capability of the network, the convolution layers are set as a first convolution layer and a second convolution layer, the size of a convolution kernel in the first convolution layer is 3 x 3, and the step length is 2; the convolution kernel in the second convolution layer has a size of 3 x 3 and a step size of 1.
The image processing in S21 is: the input image is 224 × 3, the convolution kernel size is 3 × 3 through the first convolution layer, the step size is 2, 64 features are output, and the image size is 112 × 112; after passing through the second convolution layer, the convolution kernel size is 3 × 3, the step size is 1, 64 features are output, and the image size is 112 × 112; after passing through the pooling layer with step 2, 64 features were output with an image size of 56 x 56.
And S22, extracting the characteristics of the color of the plastic bottle through a residual error network. In the residual network, in order to improve accuracy and reduce the amount of calculation, four residual blocks are provided, which are denoted as RB1, RB2, RB3, and RB 4. Each residual block has three convolution layers, namely, the network dimension of an input layer is reduced firstly, then the characteristics are learned, and finally the network dimension is restored. Wherein the three convolutional layers are respectively a third convolutional layer, a fourth convolutional layer and a fifth convolutional layer. The convolution kernel of the third convolution layer is 1 x 1 with a step size of 2, the convolution kernel of the fourth convolution layer is 3 x 3 with a step size of 1, and the convolution kernel of the fifth convolution layer is 1 x 1 with a step size of 1.
The output calculation method of the residual block is as follows:
assuming that the input of the residual block is X and the output is Y, the weight matrix of each convolution layer is W1，W2，W3The excitation function of the output of the third layer of convolutional layer is relu1The excitation function of the fourth layer convolution layer output is relu2The excitation function of the residual block output is relu3. In addition, in order to keep the consistency of the input dimension and the output dimension, assuming that the linear transformation matrix is Wt, the following output is output:
Y＝relu3([relu2(relu1(X,W1),W2)]*W3+X*Wt)
the image processing in S22 is: outputting 256 features with an image size of 56 × 56 through a residual block RB 1; outputting 512 features through a residual block RB2, wherein the image size is 28 × 28; through a residual block RB3, 1024 features are output, and the image size is 14 × 14; after passing through the residual block RB4, 2048 features were output with an image size of 7 × 7.
And S23, inputting the extracted feature map into a global average pooling layer, and inputting the pooled features into a full-connection layer for plastic bottle color classification. And carrying out global average pooling operation on 2048 7 × 7 features output by the RB4 residual block, outputting 2048 1 × 1 pooled features, inputting the pooled features to a full-connection layer, and finally obtaining a classification result.
And S24, setting parameters in the training algorithm, including learning rate, training times and iteration times.
And S25, inputting the training image data set into the established network model, and determining the deep learning network model and the model parameters. The method comprises the following specific steps:
s251, judging whether the set training times is reached, if so, obtaining a result, otherwise, repeating the S251;
s252, judging whether the set iteration times are reached, if so, finishing the training, otherwise, repeating S251;
and S253, determining a deep learning network model and model parameters.
And S3, inputting the verification image data set into the determined deep learning network model, and checking the deep learning network model.
And S4, inputting the image data set to be detected into the deep learning network model to obtain the plastic bottle color classification result in the image data set to be detected.
The above description is only for the preferred embodiment of the present invention, but the present invention should not be limited to the embodiment and the disclosure of the drawings, and therefore, all equivalent or modifications that do not depart from the spirit of the present invention are intended to fall within the scope of the present invention.
Claims (8)
1. A plastic bottle color sorting method based on image recognition is characterized by comprising the following steps:
s1, collecting a plastic bottle image, and after carrying out denoising, averaging, defogging and cutting operations on the image, dividing the image into a training image data set, a verification image data set and an image data set to be detected according to a proportion;
s2, establishing a deep learning network model for plastic bottle color sorting, inputting a training image data set into the network model for training, and determining the deep learning network model and model parameters;
s3, inputting the verification image data set into the determined deep learning network model, and checking the deep learning network model;
and S4, inputting the image data set to be detected into the deep learning network model to obtain the plastic bottle color classification result in the image data set to be detected.
2. The method for sorting colors of plastic bottles based on image recognition according to claim 1, wherein said S1 comprises the following steps;
s11, carrying out image acquisition on plastic bottles with different colors by using an industrial camera to form a first data set;
s12, carrying out Gaussian filtering denoising on the image in the first data set;
s13, carrying out gray level averaging and defogging operation processing on the image processed in the S12;
s14, cutting the graph processed in the S13 according to the size of 224 × 224, storing the cut image into a first data set, replacing the original image and forming a second data set;
s15, classifying the plastic bottles according to red, green, blue, purple, yellow, black, white and transparent colors for the image of the second data set in the S14, labeling, rotating in different directions, mirroring, increasing contrast and the like, and then, carrying out the following operations on the second data set according to the ratio of 8: 1: 1, and obtaining a training image data set, a verification image data set and an image data set to be detected after preprocessing.
3. The method for sorting colors of plastic bottles based on image recognition according to claim 2, wherein said S2 comprises the following steps;
s21, determining a convolutional layer and a pooling layer of the input image;
s22, extracting the characteristics of the color of the plastic bottle through a residual error network;
s23, inputting the extracted feature map into a global average pooling layer, and inputting the pooling features into a full-connection layer for plastic bottle color classification;
s24, setting parameters in the training algorithm, including learning rate, training times and iteration times;
and S25, inputting the training image data set into the established network model, and determining the deep learning network model and the model parameters.
4. The plastic bottle color sorting method based on image recognition according to claim 3, wherein the convolutional layers in S21 comprise a first convolutional layer and a second convolutional layer, the size of the convolutional kernel in the first convolutional layer is 3 x 3, and the step size is 2; the convolution kernel in the second convolution layer has a size of 3 x 3 and a step size of 1.
5. The method for sorting colors of plastic bottles based on image recognition according to claim 3, wherein said residual network in S22 is provided with four residual blocks, and said residual blocks are provided with a third layer of convolutional layers, a fourth layer of convolutional layers and a fifth layer of convolutional layers.
6. The method for sorting the colors of the plastic bottles based on the image recognition as claimed in claim 5, wherein the output of the residual block is calculated as follows:
Y＝relu3([relu2(relu1(X,W1),W2)]*W3+X*Wt)
wherein X is the input, Y is the output, W1、W2、W3For each convolutional layer, relu1For the excitation function output by the third convolutional layer, relu2For the excitation function output by the fourth convolution layer, relu3Wt is the linear transformation matrix for the excitation function of the residual block output.
7. The method for sorting plastic bottle colors based on image recognition according to claim 4, wherein the convolution kernel of the third convolution layer is 1 x 1 with a step size of 2, the convolution kernel of the fourth convolution layer is 3 x 3 with a step size of 1, and the convolution kernel of the fifth convolution layer is 1 x 1 with a step size of 1.
8. The method for sorting colors of plastic bottles based on image recognition as claimed in claim 3, wherein said S25 comprises the following steps:
s251, judging whether the set training times is reached, if so, obtaining a result, otherwise, repeating the S251;
s252, judging whether the set iteration times are reached, if so, finishing the training, otherwise, repeating S251;
and S253, determining a deep learning network model and model parameters.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911130056.8A CN110866561A (en) | 2019-11-18 | 2019-11-18 | Plastic bottle color sorting method based on image recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911130056.8A CN110866561A (en) | 2019-11-18 | 2019-11-18 | Plastic bottle color sorting method based on image recognition |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110866561A true CN110866561A (en) | 2020-03-06 |
Family
ID=69655412
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911130056.8A Pending CN110866561A (en) | 2019-11-18 | 2019-11-18 | Plastic bottle color sorting method based on image recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110866561A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111461298A (en) * | 2020-03-26 | 2020-07-28 | 广西电网有限责任公司电力科学研究院 | Convolutional neural network and method for circuit breaker fault identification |
| CN111583130A (en) * | 2020-04-14 | 2020-08-25 | 佛山市南海区广工大数控装备协同创新研究院 | Method for recovering license plate image for LPR |
| CN111652214A (en) * | 2020-05-26 | 2020-09-11 | 佛山市南海区广工大数控装备协同创新研究院 | Garbage bottle sorting method based on deep learning |
| CN112016462A (en) * | 2020-08-28 | 2020-12-01 | 佛山市南海区广工大数控装备协同创新研究院 | Recovery bottle classification method based on deep learning model |
| CN113592824A (en) * | 2021-08-02 | 2021-11-02 | 合肥名德光电科技股份有限公司 | Color sorter sorting method based on deep learning |
| CN114492633A (en) * | 2022-01-25 | 2022-05-13 | 合肥学院 | Colored transparent plastic classification method and system based on multi-wavelength transmission characteristic |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107292333A (en) * | 2017-06-05 | 2017-10-24 | 浙江工业大学 | A kind of rapid image categorization method based on deep learning |
| CN109190691A (en) * | 2018-08-20 | 2019-01-11 | 小黄狗环保科技有限公司 | The method of waste drinking bottles and pop can Classification and Identification based on deep neural network |
| CN110287882A (en) * | 2019-06-26 | 2019-09-27 | 北京林业大学 | A kind of big chrysanthemum kind image-recognizing method based on deep learning |
| AU2019101149A4 (en) * | 2019-09-30 | 2019-10-31 | Hu, Yaowen MR | An Image retrieval System for Brand Logos Based on Deep Learning |
-
2019
- 2019-11-18 CN CN201911130056.8A patent/CN110866561A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107292333A (en) * | 2017-06-05 | 2017-10-24 | 浙江工业大学 | A kind of rapid image categorization method based on deep learning |
| CN109190691A (en) * | 2018-08-20 | 2019-01-11 | 小黄狗环保科技有限公司 | The method of waste drinking bottles and pop can Classification and Identification based on deep neural network |
| CN110287882A (en) * | 2019-06-26 | 2019-09-27 | 北京林业大学 | A kind of big chrysanthemum kind image-recognizing method based on deep learning |
| AU2019101149A4 (en) * | 2019-09-30 | 2019-10-31 | Hu, Yaowen MR | An Image retrieval System for Brand Logos Based on Deep Learning |
Non-Patent Citations (3)
| Title |
|---|
| 刘政林等: "基于计算机视觉的塑料瓶颜色识别和分拣系统", 《电子测量技术》 * |
| 温峥嵘等: "回收塑料瓶的机器视觉颜色检测系统", 《福建电脑》 * |
| 王文君等: "基于RGB法的回收塑料瓶自动颜色分选系统开发", 《仪表技术与传感器》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111461298A (en) * | 2020-03-26 | 2020-07-28 | 广西电网有限责任公司电力科学研究院 | Convolutional neural network and method for circuit breaker fault identification |
| CN111583130A (en) * | 2020-04-14 | 2020-08-25 | 佛山市南海区广工大数控装备协同创新研究院 | Method for recovering license plate image for LPR |
| CN111652214A (en) * | 2020-05-26 | 2020-09-11 | 佛山市南海区广工大数控装备协同创新研究院 | Garbage bottle sorting method based on deep learning |
| CN111652214B (en) * | 2020-05-26 | 2024-05-28 | 佛山市南海区广工大数控装备协同创新研究院 | Garbage bottle sorting method based on deep learning |
| CN112016462A (en) * | 2020-08-28 | 2020-12-01 | 佛山市南海区广工大数控装备协同创新研究院 | Recovery bottle classification method based on deep learning model |
| CN113592824A (en) * | 2021-08-02 | 2021-11-02 | 合肥名德光电科技股份有限公司 | Color sorter sorting method based on deep learning |
| CN114492633A (en) * | 2022-01-25 | 2022-05-13 | 合肥学院 | Colored transparent plastic classification method and system based on multi-wavelength transmission characteristic |
| CN114492633B (en) * | 2022-01-25 | 2022-10-28 | 合肥学院 | Colored transparent plastic classification method and system based on multi-wavelength transmission characteristic |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110866561A (en) | Plastic bottle color sorting method based on image recognition | |
| CN108562589B (en) | Method for detecting surface defects of magnetic circuit material | |
| CN110765941B (en) | Seawater pollution area identification method and equipment based on high-resolution remote sensing image | |
| CN105447851B (en) | The sound hole defect inspection method and system of a kind of glass panel | |
| CN114627383B (en) | Small sample defect detection method based on metric learning | |
| CN112132196B (en) | Cigarette case defect identification method combining deep learning and image processing | |
| CN104408449B (en) | Intelligent mobile terminal scene literal processing method | |
| CN113298757A (en) | Metal surface defect detection method based on U-NET convolutional neural network | |
| CN111582294A (en) | Method for constructing convolutional neural network model for surface defect detection and application thereof | |
| CN110399840B (en) | Rapid lawn semantic segmentation and boundary detection method | |
| CN110766689A (en) | Method and device for detecting article image defects based on convolutional neural network | |
| CN110084302B (en) | Crack detection method based on remote sensing image | |
| CN112767369A (en) | Defect identification and detection method and device for small hardware and computer readable storage medium | |
| CN114581782B (en) | Fine defect detection method based on coarse-to-fine detection strategy | |
| CN109615604A (en) | Accessory appearance flaw detection method based on image reconstruction convolutional neural networks | |
| CN108181316A (en) | A kind of bamboo strip defect detection method based on machine vision | |
| CN113256624A (en) | Continuous casting round billet defect detection method and device, electronic equipment and readable storage medium | |
| CN104700417A (en) | Computer image based automatic identification method of timber knot flaws | |
| CN111161213A (en) | Industrial product defect image classification method based on knowledge graph | |
| CN114005081A (en) | Intelligent detection device and method for foreign matters in tobacco shreds | |
| CN115018790A (en) | Workpiece surface defect detection method based on anomaly detection | |
| CN105138975A (en) | Human body complexion area segmentation method based on deep belief network | |
| CN103065296B (en) | High-resolution remote sensing image residential area extraction method based on edge feature | |
| CN114092456A (en) | Cell fluorescence image distinguishing method and system | |
| CN110349119B (en) | Pavement disease detection method and device based on edge detection neural network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200306 |
|
| RJ01 | Rejection of invention patent application after publication |