Standard meal identification method and system based on INT4 quantization
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 kind of food and standard technology, applied in the field of neural network models, can solve the problems of high complexity of deep learning models, high hardware cost and power consumption, high hardware requirements, etc., to reduce hardware memory and computing resource requirements, large compression ratio, strong economy benefit effect
Pending Publication Date: 2022-08-09
SHANDONG INSPUR SCI RES INST CO LTD
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
However, due to the high complexity of commonly used deep learning models for image classification, the hardware requirements are high, resulting in high hardware costs and power consumption.
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
[0033] A standard meal identification method based on INT4 quantification, the specific steps of the method are as follows:
[0034] S1 collects and annotates standard meal image data to construct training and calibration datasets;
[0035] S2 uses floating-point numbers to train the image classification model to obtain the floating-point number model to be quantized;
[0036] S3 uses cross-layer equalization to preprocess the image classification model; quantizes the floating point model into an INT4 type model;
[0037] S4 deploys the INT4 quantitative model to the MCU;
[0038] S5 integrates MCU camera data as the input of the quantitative model;
[0039] Further, described S2 uses floating-point number training image classification model MobileNet-V1, obtains the floating-point number model to be quantized
[0040] Further, the S3 uses cross-layer equalization to preprocess the image classification model; the specific steps of quantizing the floating-point number model ...
Embodiment 2
[0052] A standard meal identification system based on INT4 quantification, the system specifically includes a data collection module, a model training module, a model processing module, a model deployment module and a data integration module:
[0053] Data collection module: collect and label standard meal image data, construct training and calibration datasets;
[0054] Model training module: use floating-point numbers to train image classification models to obtain floating-point numbers to be quantized;
[0055] Model processing module: use cross-layer equalization to preprocess the image classification model; quantify the floating point model into an INT4 type model;
[0056] Model deployment module: deploy the INT4 quantitative model to the MCU;
[0057] Data integration module: Integrate MCU camera data as input for quantitative model;
[0058]Further, the model training module uses the floating-point training image classification model MobileNet-V1 to obtain the floati...
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 standard meal identification method and system based on INT4 quantization, and belongs to the field of neural network models. The method comprises the following specific steps: S1, collecting and labeling standard meal image data, and constructing a training and calibration data set; s2, training an image classification model by using a floating-point number to obtain a floating-point number model to be quantized; s3, preprocessing the image classification model by using cross-layer equalization; quantizing the floating-point number model into an INT4 type model; s4, deploying an INT4 quantitative model to an MCU (Microprogrammed Control Unit); s5, integrating MCU camera data as quantitative model input; according to the method, the pipeline quantification neural network model is used, the precision loss of quantification of the neural network model can be effectively reduced, the simple standard meal identification requirement can be met, and the practicability is high; in addition, due to the fact that INT4 quantification has a large model compression proportion, the requirements for hardware memory and computing resources are reduced, and the method can be applied to a single-chip microcomputer with low cost and has high economic benefits.
Description
technical field [0001] The invention discloses a standard meal identification method and system based on INT4 quantification, and relates to the technical field of neural network models. Background technique [0002] In recent years, neural network models have been widely used in many fields and achieved very good results. However, due to the high model complexity and large size of the neural network model, the inference efficiency is low and the inference time is long. Therefore, how to design a model with low resource consumption, which can predict in real time and at the same time ensure the prediction accuracy has become a practical problem, especially in the application of simple scenarios. Standard meal recognition is a simple scene image recognition application, because the meals are relatively standard and there are relatively few types of meals, the image recognition accuracy rate is high, and it has high application value. However, due to the high complexity of t...
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