error correctionIn this paper a railway track multi-objective segmentation network based on Transformer is proposed First of all this paper constructs the railway track component datasets da-taset #1

In this paper, we propose a railway track multi-objective segmentation network based on Transformer. Firstly, we construct the railway track component datasets (dataset #1 and dataset #2). Then, we introduce an end-to-end instance segmentation model called RTMOSeg, which is trained and tested on the constructed dataset. The RTMOSeg is composed of an encoder and a decoder, where CNN and Transformer are employed in the encoder to extract global and local feature information from the input image. The decoder incorporates a multi-scale feature fusion module to enhance the segmentation performance by fusing multi-layer information. Numerous experiments conducted on the created datasets demonstrate that the proposed RTMOSeg effectively and accurately segments railway track multi-target images. This segmentation network offers a better approach and concept for the practical application of railway segmentation and detection, thereby improving the intelligence of multi-target segmentation and subsequent defect detection of rail transit components.

In the future, our first goal is to expand the railway track multi-target image dataset. Additionally, we aim to optimize the inference speed of the segmentation model proposed in this paper to make it faster