CN117636026B - Container lock pin type picture identification method - Google Patents
Container lock pin type picture identification method Download PDFInfo
- Publication number
- CN117636026B CN117636026B CN202311618851.8A CN202311618851A CN117636026B CN 117636026 B CN117636026 B CN 117636026B CN 202311618851 A CN202311618851 A CN 202311618851A CN 117636026 B CN117636026 B CN 117636026B
- Authority
- CN
- China
- Prior art keywords
- picture
- standard material
- lock pin
- standard
- target
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 20
- 239000010421 standard material Substances 0.000 claims abstract description 95
- 239000013598 vector Substances 0.000 claims abstract description 68
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 23
- 238000007781 pre-processing Methods 0.000 claims abstract description 16
- 238000000605 extraction Methods 0.000 claims abstract description 8
- 238000004364 calculation method Methods 0.000 claims description 16
- 238000011176 pooling Methods 0.000 claims description 12
- 230000004927 fusion Effects 0.000 claims description 9
- 238000001914 filtration Methods 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 6
- 230000006870 function Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/42—Global feature extraction by analysis of the whole pattern, e.g. using frequency domain transformations or autocorrelation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Medical Informatics (AREA)
- Databases & Information Systems (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- Molecular Biology (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to a method for identifying the type of a lock pin of a container, belonging to the technical field of computer vision. Wherein the method comprises the following steps: and acquiring a standard material data set and a container lock pin sample picture, and preprocessing the container lock pin sample picture to obtain a standard material. And acquiring a container lock pin acquisition picture, and preprocessing the container lock pin acquisition picture to obtain a target picture. And inputting the standard material and the target picture into a CNN convolutional neural network model, and outputting a standard material feature vector and a target picture feature vector. And calculating the similarity between the target picture and the standard material according to the target picture feature vector and the standard material feature vector. And sending the category label of the standard material corresponding to the maximum similarity value to a central control center. Feature extraction and delivery between levels is enhanced using CNN-based convolutional neural network models. The picture type of the lock pin of the container can be accurately and efficiently identified, and automatic transportation of the container is realized.
Description
Technical Field
The invention belongs to the technical field of computer vision, and particularly relates to a container type picture identification method.
Background
The container is used as a logistics transport means and is widely applied to various industries such as logistics, transportation, customs, ports and the like. The locking pins are important tools for binding the containers, the locking pins are required to be disassembled and assembled during shipping and unloading of the containers, and the locking pin type identification can realize rapid and accurate classification and scheduling of the containers, so that the logistics efficiency is improved. The lock pin type recognition belongs to the picture recognition technology, mainly applies the deep learning technology, is widely applied nowadays, can automatically extract image features, and has higher classification accuracy and robustness. The automatic identification of the type of the lock pin of the container can be realized, and the automatic disassembly of the lock pin of the container can be realized, so that the automatic transportation of the container is realized. Therefore, the lock pins are classified and identified by the picture identification technology, and the method has important significance in the aspects of improving logistics efficiency, enhancing safety, reducing cost and the like. The traditional manual identification lock pin type disassembling lock pin has low efficiency, high labor cost and low traditional automatic lock pin type identification accuracy.
Disclosure of Invention
Aiming at the pure problems in the prior art, the invention provides a container type identification method, and in the first aspect, the method can be implemented by the following technical scheme:
S1: obtaining a sample picture of a container lock pin, preprocessing the sample picture of the container lock pin to obtain standard materials, and calculating a formula according to LBP values of pixel points: Obtaining LBP values of the standard material pixel points, constructing a texture feature histogram according to the standard material LBP values to obtain texture features, carrying out distributed processing on the texture features to obtain feature vectors, carrying out weight calculation on the texture feature vectors to obtain weighted feature vectors, and classifying the weighted feature vectors to obtain a standard material data set, wherein the preprocessing comprises image normalization, filtering denoising and image enhancement;
S3: inputting the standard materials and the target pictures into a CNN convolutional neural network model, and outputting standard material feature vectors and target picture feature vectors, wherein the CNN convolutional neural network model comprises a convolutional layer, a pooling layer and a full-connection layer;
S4: calculating the similarity between the target picture and the standard material according to the target picture feature vector and the standard material feature vector;
s5: and repeating the steps S3-S4 on the standard material data set to obtain the maximum value of the similarity between the target picture and the standard material. And sending the category label of the standard material corresponding to the maximum value to a central control center.
Specifically, the preprocessing in S1 includes:
S101: normalizing pixel values of the container lock pin picture to a range of 0-1, wherein the image normalization method is to subtract the average value of the pixel values and divide the average value by a standard deviation;
s102: removing noise of a sample picture of the lock pin of the container by using a mean filtering method;
S103: the image enhancement comprises rotation, clipping and scaling, wherein the container lock pin picture is clipped to only a container lock pin part, the container lock pin picture is scaled to 200 multiplied by 200 pixels, and the container lock pin picture is rotated by 90 degrees.
Specifically, the operation of the convolution layer in S3 includes:
And inputting the standard materials and the target pictures into the convolution layer, and carrying out texture feature extraction, angular point feature extraction and edge feature extraction on the standard materials and the target pictures to obtain a standard feature map and a target feature map.
Specifically, the operation of the pooling layer in S3 includes:
inputting the standard feature map and the target feature map into the pooling layer to obtain a standard dimension reduction feature map and a target dimension reduction feature map;
Specifically, the operation of the full connection layer in S3 includes:
and inputting the standard dimension reduction feature map and the target dimension reduction feature map into a full connection layer to obtain standard material feature vectors and target picture feature vectors.
Specifically, the step S4 is:
s401: calculating cosine similarity of the standard material feature vector and the target picture feature vector;
s402: and calculating a weighted value according to the cosine similarity, wherein the calculation formula is as follows: J is the total number of standard materials in the standard material data set, and a weighted standard material characteristic vector and a weighted target picture characteristic vector are obtained according to the weighted value;
s403: and fusing the weighted standard material characteristics and the weighted target picture characteristics to obtain fusion characteristics, wherein a fusion formula is as follows: v= [ x; y; x+y; and (3) calculating probability distribution of the fusion feature by using a softmax classifier, and obtaining the similarity u of the standard material and the target picture according to the probability distribution.
Specifically, the loss function of the model in S3 is defined as:
wherein z is a class label of the standard material, and u is the similarity between the target picture and the standard material.
In a second aspect, the invention provides a container pin class identification system, operating as claimed above, comprising the following modules:
the system comprises a data set, an acquisition picture acquisition module, a vector generation module, a similarity calculation module and a central control center, wherein the standard material data set is a correlation data set of a pre-constructed container lock pin sample picture and a category label;
the acquisition picture acquisition module is used for acquiring the acquisition picture of the container lock pin, and preprocessing the acquisition picture of the container lock pin to obtain a target picture;
The vector generation module comprises the steps that standard materials and target pictures are input into the CNN convolutional neural network, and are input into the full-connection layer after being subjected to convolution for 3 times and pooling to obtain standard material feature vectors and target picture feature vectors;
the similarity calculation module is used for calculating the similarity of the standard material feature vector and the target picture feature vector to obtain the similarity of the standard material and the target picture;
the central control center is used for receiving the category labels corresponding to the standard materials when the similarity is maximum.
The beneficial effects of the invention are as follows:
(1) The deep convolutional neural network is used, feature extraction of the lock pin picture of the container is enhanced, feature loss is reduced as much as possible, and the class identification accuracy is high and the efficiency is high.
(2) Through the image recognition technology, intelligent management and monitoring of the lock pin of the container can be realized, so that the transportation cost is reduced and the waste of human resources is reduced.
Drawings
The present invention is further described below with reference to the accompanying drawings for the convenience of understanding by those skilled in the art.
FIG. 1 is a flow chart of the container lock pin class image recognition method.
Fig. 2 is a block diagram of a container lock pin class based image recognition system according to the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the invention for achieving the preset aim, the following detailed description is given below of the specific implementation, structure, characteristics and effects according to the invention with reference to the attached drawings and the preferred embodiment.
Referring to fig. 1, a flow chart of a container lock pin type picture recognition method includes the following steps:
S1: obtaining a sample picture of a container lock pin, preprocessing the sample picture of the container lock pin to obtain standard materials, and calculating a formula according to LBP values of pixel points: obtaining LBP values of the standard material pixel points, constructing a texture feature histogram according to the standard material LBP values to obtain texture features, carrying out distributed processing on the texture features to obtain feature vectors, carrying out weight calculation on the texture feature vectors to obtain weighted feature vectors, and classifying the weighted feature vectors to obtain a standard material data set, wherein the preprocessing comprises image normalization, filtering denoising and image enhancement;
S2: acquiring a container lock pin acquisition picture, and preprocessing the container lock pin acquisition picture to obtain a target picture;
S3: inputting the standard materials and the target pictures into a CNN convolutional neural network model, and outputting standard material feature vectors and target picture feature vectors, wherein the CNN convolutional neural network model comprises a convolutional layer, a pooling layer and a full-connection layer;
S4: calculating the similarity between the target picture and the standard material according to the target picture feature vector and the standard material feature vector;
s5: and repeating the steps S3-S4 on the standard material data set to obtain the maximum value of the similarity between the target picture and the standard material. And sending the category label of the standard material corresponding to the maximum value to a central control center.
Specifically, the preprocessing in S1 includes:
S101: normalizing pixel values of the container lock pin picture to a range of 0-1, wherein the image normalization method is to subtract the average value of the pixel values and divide the average value by a standard deviation;
s102: removing noise of a sample picture of the lock pin of the container by using a mean filtering method;
S103: the image enhancement comprises rotation, clipping and scaling, wherein the container lock pin picture is clipped to only a container lock pin part, the container lock pin picture is scaled to 200 multiplied by 200 pixels, and the container lock pin picture is rotated by 90 degrees.
Specifically, the operation of the convolution layer in S3 includes:
Inputting the standard materials and the target pictures into the convolution layer, and extracting texture features, corner features and edge features of the standard materials and the target pictures to obtain a standard feature map and a target feature map, wherein the calculation formula is as follows: In/> Input feature map for ith layer of layer 1,/>A convolution kernel from the ith input characteristic image of the first layer-1 to the jth output characteristic image of the first layer; /(I)And outputting the characteristic diagram for the j-th sheet of the first layer.
Specifically, the operation of the pooling layer in S3 includes:
inputting the standard feature map and the target feature map into the pooling layer to obtain a standard dimension reduction feature map and a target dimension reduction feature map, wherein a calculation formula is as follows: wherein: down () is a downsampling function; beta is the network multiplicative bias;
Specifically, the operation of the full connection layer in S3 includes:
and inputting the standard dimension reduction feature map and the target dimension reduction feature map into a full connection layer to obtain standard material feature vectors and target picture feature vectors.
Specifically, the calculation method of the similarity between the standard material and the target picture in S4 is as follows:
s401: and calculating cosine similarity of the standard material feature vector and the target picture feature vector, wherein a calculation formula is as follows: Wherein/> For the standard material feature vector,/>The target picture feature vector is;
s402: and calculating a weighted value according to the cosine similarity, wherein the calculation formula is as follows: J is the total number of standard materials in the standard material data set, and a weighted standard material characteristic vector and a weighted target picture characteristic vector are obtained according to the weighted value;
s403: and fusing the weighted standard material characteristics and the weighted target picture characteristics to obtain fusion characteristics, wherein a fusion formula is as follows: v= [ x; y; x+y; and (3) calculating probability distribution of the fusion feature by using a softmax classifier, and obtaining the similarity u of the standard material and the target picture according to the probability distribution.
Specifically, the loss function of the model in S3 is defined as:
wherein z is a class label of the standard material, and u is the similarity between the target picture and the standard material.
In a second aspect, the invention provides a container pin class identification system, operating as claimed above, comprising the following modules:
the system comprises a data set, an acquisition picture acquisition module, a vector generation module, a similarity calculation module and a central control center, wherein the standard material data set is a correlation data set of a pre-constructed container lock pin sample picture and a category label;
the acquisition picture acquisition module is used for acquiring the acquisition picture of the container lock pin, and preprocessing the acquisition picture of the container lock pin to obtain a target picture;
The vector generation module comprises the steps that standard materials and target pictures are input into the CNN convolutional neural network, and are input into the full-connection layer after being subjected to convolution for 3 times and pooling to obtain standard material feature vectors and target picture feature vectors;
the similarity calculation module is used for calculating the similarity of the standard material feature vector and the target picture feature vector to obtain the similarity of the standard material and the target picture;
the image acquisition module is used for acquiring a sample picture of the container lock pin and an acquisition picture of the container lock pin. The present invention is not limited to the above embodiments, but is capable of modification and variation in detail, and other modifications and variations can be made by those skilled in the art without departing from the scope of the present invention.
Claims (7)
1. The picture identification method of the lock pin class of the container is characterized by comprising the following steps:
S1: obtaining a sample picture of a container lock pin, preprocessing the sample picture of the container lock pin to obtain standard materials, and calculating a formula according to LBP values of pixel points: Obtaining LBP values of the standard material pixel points, constructing a texture feature histogram according to the standard material LBP values to obtain texture features, carrying out distributed processing on the texture features to obtain feature vectors, carrying out weight calculation on the texture feature vectors to obtain weighted feature vectors, and classifying the weighted feature vectors to obtain a standard material data set, wherein the preprocessing comprises image normalization, filtering denoising and image enhancement;
S2: acquiring a container lock pin acquisition picture, and preprocessing the container lock pin acquisition picture to obtain a target picture;
S3: inputting the standard materials and the target pictures into a CNN convolutional neural network model, and outputting standard material feature vectors and target picture feature vectors, wherein the CNN convolutional neural network model comprises a convolutional layer, a pooling layer and a full-connection layer;
S4: according to the characteristic vector of the target picture and the characteristic vector of the standard material, calculating the similarity of the target picture and the standard material, wherein the specific steps comprise:
s401: calculating cosine similarity of the standard material feature vector and the target picture feature vector;
s402: and calculating a weighted value according to the cosine similarity, wherein the calculation formula is as follows: Wherein/> Obtaining weighted standard material feature vectors and weighted target picture feature vectors for the total number of standard materials in the standard material data set according to the weighted values;
S403: and fusing the weighted standard material characteristics and the weighted target picture characteristics to obtain fusion characteristics, wherein a fusion formula is as follows: Wherein/> For the standard material feature,/>For target picture features, calculating probability distribution of the fusion features by using a softmax classifier, and obtaining the similarity/>, of the standard materials and the target picture, according to the probability distribution；
S5: repeating the steps S3-S4 on the standard material data set to obtain the maximum value of the similarity between the target picture and the standard material; and sending the category label of the standard material corresponding to the maximum value to a central control center.
2. The method according to claim 1, wherein the preprocessing in S1 comprises:
S101: normalizing pixel values of the container lock pin picture to a range of 0-1, wherein the image normalization method is to subtract the average value of the pixel values and divide the average value by a standard deviation;
s102: removing noise of a sample picture of the lock pin of the container by using a mean filtering method;
S103: the image enhancement comprises rotation, clipping and scaling, wherein the container lock pin picture is clipped to only comprise the container lock pin part, the container lock pin picture is scaled to 200 multiplied by 200 pixels, and the container lock pin picture is rotated by 90 degrees.
3. The method of claim 1, wherein the operation of the convolution layer in S3 comprises:
And inputting the standard materials and the target pictures into the convolution layer, and carrying out texture feature extraction, angular point feature extraction and edge feature extraction on the standard materials and the target pictures to obtain a standard feature map and a target feature map.
4. A method according to claim 3, wherein the operation of the pooling layer in S3 comprises:
And inputting the standard feature map and the target feature map into the pooling layer to obtain a standard dimension reduction feature map and a target dimension reduction feature map.
5. The method of claim 4, wherein the operation of the full connectivity layer in S3 comprises:
and inputting the standard dimension reduction feature map and the target dimension reduction feature map into a full connection layer to obtain standard material feature vectors and target picture feature vectors.
6. The method according to claim 1, wherein the loss function of the model in S3 is defined as: Wherein z is a class label of the standard material,/> And the similarity between the target picture and the standard material is obtained.
7. A system for identifying a shipping container lock pin category picture, operating with the method of any one of claims 1-6, comprising the following modules:
The system comprises a standard material data set, a target picture acquisition module, a vector generation module, a similarity calculation module and a central control center, wherein the standard material data set is a pre-constructed association data set of a container lock pin sample picture and a category label;
The target picture acquisition module is used for acquiring the picture acquired by the container lock pin, and preprocessing the picture acquired by the container lock pin to obtain a target picture;
The vector generation module inputs the standard material and the target picture into the CNN convolutional neural network, and inputs the standard material feature vector and the target picture feature vector into the full connection layer after convolution is performed for 3 times and pooling is performed, so that the standard material feature vector and the target picture feature vector are obtained;
the similarity calculation module is used for calculating the similarity of the standard material feature vector and the target picture feature vector to obtain the similarity of the standard material and the target picture;
the central control center is used for receiving the category labels corresponding to the standard materials when the similarity is maximum.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311534756X | 2023-11-17 | ||
| CN202311534756 | 2023-11-17 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117636026A CN117636026A (en) | 2024-03-01 |
| CN117636026B true CN117636026B (en) | 2024-06-11 |
Family
ID=90017739
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311618851.8A Active CN117636026B (en) | 2023-11-17 | 2023-11-30 | Container lock pin type picture identification method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117636026B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019237646A1 (en) * | 2018-06-14 | 2019-12-19 | 清华大学深圳研究生院 | Image retrieval method based on deep learning and semantic segmentation |
| CN110751027A (en) * | 2019-09-09 | 2020-02-04 | 华中科技大学 | Pedestrian re-identification method based on deep multi-instance learning |
| CN112287144A (en) * | 2020-10-29 | 2021-01-29 | 苏州科达科技股份有限公司 | Picture retrieval method, equipment and storage medium |
| WO2023206944A1 (en) * | 2022-04-25 | 2023-11-02 | 深圳思谋信息科技有限公司 | Semantic segmentation method and apparatus, computer device, and storage medium |
-
2023
- 2023-11-30 CN CN202311618851.8A patent/CN117636026B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019237646A1 (en) * | 2018-06-14 | 2019-12-19 | 清华大学深圳研究生院 | Image retrieval method based on deep learning and semantic segmentation |
| CN110751027A (en) * | 2019-09-09 | 2020-02-04 | 华中科技大学 | Pedestrian re-identification method based on deep multi-instance learning |
| CN112287144A (en) * | 2020-10-29 | 2021-01-29 | 苏州科达科技股份有限公司 | Picture retrieval method, equipment and storage medium |
| WO2023206944A1 (en) * | 2022-04-25 | 2023-11-02 | 深圳思谋信息科技有限公司 | Semantic segmentation method and apparatus, computer device, and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117636026A (en) | 2024-03-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Bouti et al. | A robust system for road sign detection and classification using LeNet architecture based on convolutional neural network | |
| US6915025B2 (en) | Automatic image orientation detection based on classification of low-level image features | |
| CN111914835A (en) | Bill element extraction method and device, electronic equipment and readable storage medium | |
| CN111652317B (en) | Super-parameter image segmentation method based on Bayes deep learning | |
| Joshi et al. | Recent advances in local feature detector and descriptor: a literature survey | |
| Zakir et al. | Road sign detection and recognition by using local energy based shape histogram (LESH) | |
| CN109145964B (en) | Method and system for realizing image color clustering | |
| Chiu et al. | A novel directional object detection method for piled objects using a hybrid region-based convolutional neural network | |
| Agrawal et al. | Supervised approach for object identification using speeded up robust features | |
| CN112633297A (en) | Target object identification method and device, storage medium and electronic device | |
| CN112784754A (en) | Vehicle re-identification method, device, equipment and storage medium | |
| CN111414917B (en) | Identification method of low-pixel-density text | |
| WO2019071476A1 (en) | Express information input method and system based on intelligent terminal | |
| CN111178464A (en) | Application of OCR recognition based on neural network in logistics industry express bill | |
| CN108090728B (en) | Express information input method and system based on intelligent terminal | |
| CN117292193B (en) | Multi-station intelligent logistics conveying system | |
| CN117636026B (en) | Container lock pin type picture identification method | |
| Quispe et al. | Shape prior based image segmentation using manifold learning | |
| CN105844299B (en) | A kind of image classification method based on bag of words | |
| CN111476524A (en) | Goods picking method, system and device for finished product delivery based on artificial intelligence and readable storage medium | |
| CN112529513A (en) | Intelligent seal checking method and system | |
| CN115601586A (en) | Label information acquisition method and device, electronic equipment and computer storage medium | |
| Nguyen et al. | Digital transformation for shipping container terminals using automated container code recognition | |
| CN115496948A (en) | Network supervision fine-grained image identification method and system based on deep learning | |
| Chen et al. | Design and Implementation of Second-generation ID Card Number Identification Model based on TensorFlow |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |