Precise lawn and pasture weeding method based on cloud weed control spectrum

Abstract

The invention relates to a precise lawn and pasture weeding method based on a cloud weed control spectrum. The weeding method comprises the following steps: establishing a deep learning model according to a herbicide weed control spectrum, and identifying weed types based on the herbicide weed control spectrum; deploying the trained neural network model to a cloud, and performing interaction between the cloud and a weeding robot by using a 5G network; and constructing an intelligent spraying decision-making system capable of supporting multi-task parallelism, and realizing an early warning and manual intervention mechanism through multiple terminals. According to the weeding method, based on weed control spectrum identification and spraying, the herbicide can be effectively saved, the weeding effect is more accurate and efficient, the performance of weed identification can be effectively improved by means of the computing power of a cloud large-scale server and the high-speed data transmission of the 5G network, and a multi-task scene can be processed at the same time.

Description

technical field [0001] The invention relates to the field of precise weeding of lawns and pastures, in particular to a method for precise weeding of lawns and pastures based on a cloud-based herbicide spectrum. Background technique [0002] Lawns are often found in public facilities such as parks and stadiums. Lawn weed control is the key content of lawn maintenance. In our country, due to the invasion of weeds, the lawn often degenerates or even becomes deserted within a few years. For forage grass, weeding is more important because of its slow growth at the seedling stage, and the control of weeds seriously affects the quality of forage grass. Herbicide-based chemical weed control is a common method of weed control in lawns and pastures. Precise spraying of herbicides can effectively reduce the amount of herbicides and reduce environmental pollution. Weed identification is the premise and key to precise weeding. In recent years, artificial intelligence technology rela...

AI Technical Summary

Problems solved by technology

Herbicide spraying that does not differentiate herbicidal spectrum is not only a waste of herbicide, but also often futile

[0004] 2. The way of stand-alone deployment is limited by the computing power of the deployed terminal, and it is difficult to further improve the performance of the algorithm

[0005] 3. The update operation of the recognition model is cumbersome

[0006] 4. It is difficult to manually intervene in the weeding process; there is no intuitive collection and display medium for weed identification data and spraying results

Images