Mask R-CNN-based mango instance adversarial segmentation method

Abstract

The invention discloses a Mask R-CNN-based mango instance adversarial segmentation method. The method comprises the steps of establishing a mango segmentation data set in a natural scene; constructinga segmentation network based on the Mask R-CNN; regarding the constructed Mask R-CNN segmentation network as a generation network, and adding a discrimination network to a Mask branch of the Mask R-CNN segmentation network; enabling the generation network to obtain a prediction instance mask of the mango from the input image, wherein the input of the discrimination network is a real or Fake mangoinstance; replacing the original binary cross entropy of the Mask branch with the SmoothL1 + IOU Loss, wherein the generation network and the discrimination network are subjected to optimization training through an alternate confrontation strategy, so that an adversarial network model is formed; performing instance segmentation of mango fruits on the trained adversarial network model, and thus improving indexes of detection and segmentation obviously.

Description

technical field [0001] The invention relates to the technical field of fruit segmentation of agricultural computer vision, in particular to a Mango instance confrontation segmentation method based on Mask R-CNN. Background technique [0002] Mango is one of the common fruits in people's daily life. In recent years, with the continuous expansion of mango cultivation area in the world, the output has increased year by year, and mango has become one of the top five fruits in the world. Among them, China is the country with the second largest mango harvesting area in the world, accounting for 17%. Mango occupies an important position in the development of my country's fruit industry. However, agricultural labor is increasingly scarce, and it is urgent to improve the level of mechanical automation in orchards. Fruit instance segmentation is an important prerequisite for orchard machinery automation. [0003] In terms of fruit detection and segmentation, traditional machine le...

AI Technical Summary

Problems solved by technology

Finally, the improved model is used for instance segmentation of mango fruit to solve the detection and segmentation problems caused by uneven illumination of the fruit skin, occlusion of branches and leaves, fruit overlap, and too small mango target in the natural orchard scene.

Images