A multi-angle garbage classification cloud platform based on deep learning

Abstract

The invention discloses a multi-angle garbage classification cloud platform based on deep learning, and the platform comprises the steps: obtaining a garbage image through a front-end image collectionmodule, and judging whether the garbage image can meet the classification requirement or not; transmitting the garbage image to a garbage classification cloud platform through a wireless high-speed transmission module; enabling the garbage classification cloud platform to identify the garbage image, using a garbage classification module based on a deep learning model for giving a classification result, and storing each identification result in a cloud platform database module; and meanwhile, transmitting a classification result to the front-end image acquisition and transmission module, enabling the control module to classify the garbage according to the received classification command, and after garbage classification is completed, the front end transmits a confirmation instruction to the cloud platform to end the classification process. According to the method, a mode of combining the garbage classification module and the database module is adopted, a distributed system is used, upgrading and updating of the deep learning model are facilitated, and the recognition accuracy is improved by combining a multi-angle recognition and judgment technology.

Description

technical field [0001] The invention relates to a classification method and system, in particular to a multi-angle garbage classification cloud platform based on deep learning. Background technique [0002] At present, new attempts have been made in socialization and marketization of domestic waste classification in cities in our country, but the coverage is not high, and it is difficult to promote the classification work. Through waste classification management, the utilization of waste resources can be maximized, the amount of waste disposal can be reduced, and the environmental quality can be improved. To realize waste classification, the primary problem is waste classification and placement. The difficulty lies in the complexity and variety of subjects involved and the wide range of points. Relevant stakeholders include both residents and institutions, both living consumption and business activities. As far as the residents are concerned, there are not only spatial diff...

AI Technical Summary

Problems solved by technology

Secondly, the difficulty of garbage classification collection is low collection activity density, high logistics cost, and low profit of secondary sorting

Due to the lack of centralized and unified specifications and types, the waste components are complex, and it is difficult to form a scale effect for classified collection. The average cost is high, and the collection cost cannot be covered by self-viability alone.

In the case of insufficient government financial support, the classified collection of garbage is unsustainable, causing environmental pollution and wasting land resources

[0003] The existing garbage sorting system can only be sorted and identified on the on-site sorting device, which is time-consuming, costly, difficult to upgrade and maintain, and only uses a single camera to collect garbage pictures

Images