Intelligent image data collecting and processing method for refrigerator

Abstract

The invention discloses an intelligent image data collecting and processing method for a refrigerator, which comprises the following steps of: collecting image data: controlling camera equipment to collect and send the image data according to a preset triggering condition; recognizing an object, detecting the image by using an enhanced YOLOv4-tiny model, extracting the detected target object, carrying out category classification, forming a relation hierarchical graph and a matching weight matrix based on image data, and confirming the category of the object and the object through an object recognition network and the matching weight matrix; motion recognition: constructing a space-time diagram which is free of intersection and disappears along with time coverage based on the image data, and tracking an action track based on the space-time diagram; scheduling of the virtual server: scheduling the virtual server to pull and process the image data based on the collection state of the image data, the recognition state of the object and the recognition state of the action; the intelligent refrigerator control system can be applied to external camera shooting equipment, intelligent control over the interior of the external camera shooting equipment is achieved. The intelligent refrigerator control system has the intelligent image data collecting and processing functions.

Description

technical field [0001] The invention relates to an intelligent refrigerator control system. Background technique [0002] A refrigerator is a refrigeration device that maintains a constant low temperature, and it is also a civilian product that keeps food or other items at a constant low temperature. There are compressors, refrigerators or cabinets for freezing in the box, and storage boxes with refrigeration devices. The volume of household refrigerators is usually 20-500 liters. [0003] In 1910, the world's first compression refrigeration household refrigerator came out in the United States. In 1925, the Swedish Lido Company developed a household absorption refrigerator. In 1927, General Electric Company of the United States developed a fully enclosed refrigerator. In 1930, the air-cooled continuous diffusion absorption refrigerator with different heating methods was put on the market. In 1931, a new type of refrigerant Freon 12 was successfully developed. In the sec...

AI Technical Summary

Problems solved by technology

[0005] Traditional refrigerators do not have the function of visually recognizing food. At present, smart refrigerators rarely have this function and the price is relatively expensive. After buying a smart refrigerator, its functions are fixed and the volume is large. If the user wants to upgrade or replace it, the operation is very difficult. and high cost

[0006] It can be seen from the above that one of the reasons why traditional refrigerators cannot be directly upgraded to smart refrigerators is the realization of the intelligent function of visually recognizing food. More specifically, traditional refrigerators lack the component of smart camera equipment, and There is a camera on the smart refrigerator. Therefore, the smart refrigerator can collect food images through the camera and realize intelligent control through the corresponding software program system. However, due to the lack of smart camera equipment in the traditional refrigerator, it is necessary to install a camera on the outside. The equipment also lacks its internal intelligent control means, and cannot carry out intelligent image data collection and processing

Images