Unlock instant, AI-driven research and patent intelligence for your innovation.
Target identification method based on lightweight neural network and application thereof
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 target recognition technology, which is applied to the target recognition method based on lightweight neural network and its application field, can solve the problems of small data processing volume, high precision and data processing equipment hardware requirements, and achieve small data processing volume, The effect of good flexibility and high recognition accuracy
Pending Publication Date: 2021-10-15
SHANGHAI NORMAL UNIVERSITY
View PDF0 Cites 0 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
[0006] The purpose of the present invention is to overcome the existing defects that it is difficult to achieve a good balance between precision and data processing capacity, and has high requirements for equipment hardware, and to provide a target with high precision and small dataprocessing capacity, which can be applied to embedded devices with limited performance. recognition methods
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
[0045] A target recognition method based on a lightweight neural network, the steps of which are as follows:
[0046] (1) Training Densenet improved model:
[0047] Densenet improved model such as figure 1 As shown, the main convolutional layer (main Conv), feature extraction layer, first network block, transition layer, second network block and classification layer connected in sequence are included;
[0048] The first network block and the second network block have the same structure, and the network block includes the first Back Bone, the second BackBone, the third Back Bone and the fourth Back Bone connected in sequence, and the output of the first Back Bone is the second Back Bone, the second Back Bone, and the fourth Back Bone. The input of the third Back Bone and the fourth Back Bone, the output of the second Back Bone is also the input of the third Back Bone and the fourth Back Bone;
[0049] The Back Bone in the network block includes Channel Split, the first convol...
Embodiment 2
[0066] A spare kit, including a central host and five embedded devices, the embedded devices communicate with the central host;
[0067] The embedded device includes one or more processors, one or more memories, one or more programs and an input unit, the input unit is used to input target pictures, one or more programs are stored in the memory, when one or more When the program is executed by the processor, the embedded device performs the same target recognition method based on a lightweight neural network as in Embodiment 1;
[0068] The central host runs as figure 2 The program shown:
[0069] (1) The central host (based on RabbitMQ is figure 2 mq running in middle) to obtain the heartbeat packets regularly sent by each embedded device, confirm the online status of the embedded device and eliminate the offline embedded device;
[0070] (2) The central host (based on) sends the target picture to each online embedded device, and each online embedded device runs a target...
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 a target identification method based on a lightweight neural network and application thereof. The method comprises the following steps of: inputting a target picture into a trained Densenet improved model to obtain a target category; wherein the Densenet improved model is different from the Densenet model in that a Back Bone in a net block, a Channel Split in the Back Bone, a first convolutional layer, a first depth separation convolutional layer, a second convolutional layer, a Concat and a Channel Shuffle are connected in sequence, the second depth separation convolutional layer and a third convolutional layer are connected with the first convolutional layer, the first depth separation convolutional layer and the second convolutional layer in parallel, the second depth separation convolutional layer is connected with the Channel Split, and the third convolutional layer is connected with the Concat. The target identification method is small in data processing amount, high in identification precision and promising in application prospect.
Description
technical field [0001] The invention belongs to the technical field of equipment identification, and relates to a target identification method based on a lightweight neural network and an application thereof. Background technique [0002] Target detection is one of the important research directions in the field of computer vision. The traditional target detection method is to extract features by constructing feature descriptors and then use classifiers to classify features to achieve target detection, such as the histogram of gradient orientation HOG (Histogram of Oriented Gradient) And support vector machine SVM (Support Vector Machine). With the excellent performance of deep learning in the field of image classification, convolutional neural networks have been widely used in various fields of computer vision. Using deep learning to achieve target detection has become a new direction in the field of target detection. [0003] The traditional neural network uses a fully co...
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