Video content intelligent analysis method based on edge computing platform

By constructing a hierarchical quantized neural network model on the edge computing platform and dynamically scheduling the number of NPU cores, the problems of accuracy loss and resource utilization in video analysis on the edge computing platform are solved, and efficient and real-time intelligent analysis of video content is achieved.

CN121982489APending Publication Date: 2026-05-05FUJIAN ZHONGRUI NETWORK CO LTD
0 Cites 1 Cited by

Patent Information

Application Number
CN202610083271.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-01-22
Publication Date
2026-05-05

AI Technical Summary

Technical Problem

Existing edge computing platforms face problems in intelligent video content analysis, such as loss of model deployment accuracy, improper scheduling of computing resources, low efficiency of heterogeneous computing resource collaboration, and insufficient adaptability to multiple scenarios, resulting in low analysis accuracy, high latency, and poor energy efficiency.

Method used

A hierarchical quantization strategy is adopted to quantize the neural network model. A pipelined parallel architecture is built by combining the heterogeneous characteristics of CPU and NPU. The number of NPU cores is dynamically scheduled. The event-driven framework realizes the collaborative scheduling between tasks and supports the rapid switching of multiple models and exception handling.

Benefits of technology

It significantly improves the real-time performance and computing power utilization of edge video analysis, meets the analysis requirements of high precision and low latency, has good scene adaptability and flexible scalability, and reduces system power consumption.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention provides a video content intelligent analysis method based on an edge computing platform, and the method comprises the steps: collecting a video stream through an image collection interface of the edge computing platform, and carrying out the preprocessing of a video frame at a CPU end, and obtaining standardized image data; the method comprises the following steps: converting a pre-trained neural network model into a special format model adaptive to an NPU (Network Processing Unit), and adopting a hierarchical quantization strategy in the conversion process: quantizing at least one network layer responsible for basic feature extraction in the model by adopting first numerical value precision, quantizing other network layers responsible for classification and regression in the model by adopting second numerical value precision lower than the first numerical value precision; dynamically scheduling the number of processing cores in an activated state in the NPU based on a processing load determined by analyzing the current video frame content represented by the standardized image data, and utilizing the processing cores in the activated state to load a special format model to infer the standardized image data; and post-processing the reasoning output result of the NPU to generate a structured analysis result.
Need to check novelty before this filing date? Find Prior Art