An image classification method based on bi-directional neural network structure
What is Al technical title?
Al technical title is built by PatSnap Al team. It summarizes the technical point description of the patent document.
A neural network and network structure technology, applied in the field of image classification algorithms, can solve the problems of consuming a lot of manpower and material resources, cannot guarantee the classification effect, and improve it, so as to achieve obvious effects and improve the classification accuracy
Active Publication Date: 2018-12-28
以萨技术股份有限公司 +1
View PDF8 Cites 2 Cited by
Summary
Abstract
Description
Claims
Application Information
AI Technical Summary
This helps you quickly interpret patents by identifying the three key elements:
Problems solved by technology
Method used
Benefits of technology
Problems solved by technology
[0003] At present, with the advent of the big data era, how to improve the classification accuracy of the model through massive samples has become an important challenge and difficulty. Image classification algorithms have made great progress in recent years. Some algorithms believe that by increasing the width of the network It can improve the feature extraction ability and further improve the classification accuracy. The current algorithms rely on increasing the width or depth of the network to improve the performance of the model. If you want to further improve the performance of the algorithm, you need to repeatedly design and adjust the network and train the fine-tuning model, which will consume a lot of time. A lot of manpower and material resources, and there is no guarantee that the final classification effect will be improved
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more
Image
Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
Click on the blue label to locate the original text in one second.
Reading with bidirectional positioning of images and text.
Smart Image
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 1
[0048] Include the following steps:
[0049] Step S1, replace the fully connected layer of CaffeNet with a Two-Directional layer containing two transformation matrices;
[0050] Step S2, assuming that layer l is a bidirectional neural network layer, the previous layer is layer l-1, and the next layer is layer l+1. The definition of this layer can be expressed as the following formula:
[0051]
[0052] Among them, mmn l It is a new feature after the original feature is transformed by bidirectional neural network dimensionality reduction.
[0053] Step S3, therefore in the two-way neural network layer, the output of the first layer can be expressed as the following formula:
[0054]
[0055] Where f( ) represents the activation operation, and the RELU activation function f(x)=max(0,x) is used to solve the problem of gradient disappearance in the training process; where W l is the weight of layer l, b l is the bias of layer l, It is equivalent to the weight matrix W ...
Embodiment 2
[0085] Implement the Two-Direction image classification algorithm on VGGNet, using such as figure 1 The process shown includes the following steps:
[0086] Step S1, using a Two-Directional layer containing two transformation matrices to replace the fully connected layer of VGGNet;
[0087] Steps S2 to S8 are the same as in Embodiment 1.
Embodiment 3
[0089] On the Caltech-256 data set, implement Example 1 and Example 2 respectively, and compare the classification performance before and after adding the Two-Direction layer.
[0090] Table 1 Caltech-256 data set experimental results
[0091]
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more
PUM
Login to view more
Abstract
The invention discloses an image classification method based on a bi-directional neural network structure, comprising the following steps: 1. The Directional layer replaces the full-connection layer in a traditional convolution network, the image classification model based on bi-directional neural network is built, 2. network forward propagation is performed, and by adding transformation matricesL and R, the rectangular structure of the last convolution layer is preserved, steps 2, 3 are repeated, the bi-directional neural network is fine-tuned until the classification network converges. 5, the class number of the image is obtained through forward propagation on the trained model. A bidirectional neural network structure is used on the premise of keeping the depth and width of the networkconstant, the dimension of the feature is transformed by designing the transformation matrix, which ensures the matrix form of convolution, effectively preserves the structure information of the original feature space of the image, and avoids the loss of spatial structure information in the process of matrix drawing into vectors in the whole connection layer.
Description
technical field [0001] The invention belongs to the field of deep learning and artificial intelligence, in particular to an image classification algorithm, which is used for large-scale image labeling and attribute classification. Background technique [0002] Image classification is one of the important directions of artificial intelligence research. It has many applications in real life. For example, in intelligent traffic monitoring, whether vehicles are covered or not, hanging license plates, agricultural satellites determine agricultural and non-agricultural areas, and Baidu launched Baidu image recognition, Google's flower recognition, etc., if the attribute prediction and classification of pictures rely entirely on manual work, a lot of manpower, material and financial resources will be invested. Using image classification technology based on deep learning can quickly and accurately classify images. [0003] At present, with the advent of the era of big data, how to i...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more
Application Information
Patent Timeline
Application Date:The date an application was filed.
Publication Date:The date a patent or application was officially published.
First Publication Date:The earliest publication date of a patent with the same application number.
Issue Date:Publication date of the patent grant document.
PCT Entry Date:The Entry date of PCT National Phase.
Estimated Expiry Date:The statutory expiry date of a patent right according to the Patent Law, and it is the longest term of protection that the patent right can achieve without the termination of the patent right due to other reasons(Term extension factor has been taken into account ).
Invalid Date:Actual expiry date is based on effective date or publication date of legal transaction data of invalid patent.
Login to view more 
Login to view more