Semantic segmentation method of underwater images based on encoder-decoder architecture

Abstract

With the exploration and development of marine resources, deep learning is more and more widely used in underwater image processing. However, the quality of the original underwater images is so low that traditional semantic segmentation methods obtain poor segmentation results, such as blurred target edges, insufficient segmentation accuracy, and poor regional boundary segmentation effects. To solve these problems, this paper proposes a semantic segmentation method for underwater images. Firstly, the image enhancement based on multi-spatial transformation is performed to improve the quality of the original images, which is not common in other advanced semantic segmentation methods. Then, the densely connected hybrid atrous convolution effectively expands the receptive field and slows down the speed of resolution reduction. Next, the cascaded atrous convolutional spatial pyramid pooling module integrates boundary features of different scales to enrich target details. Finally, the context information aggregation decoder fuses the features of the shallow network and the deep network to extract rich contextual information, which greatly reduces information loss. The proposed method was evaluated on RUIE, HabCam UID, and UIEBD. Compared with the state-of-the-art semantic segmentation algorithms, the proposed method has advantages in segmentation integrity, location accuracy, boundary clarity, and detail in subjective perception. On the objective data, the proposed method achieves the highest MIOU of 68.3 and OA of 79.4, and it has a low resource consumption. Besides, the ablation experiment also verifies the effectiveness of our method.

Fig 1. The original underwater images with low image quality.

Fig 2. The pipeline of the proposed method.

Fig 3. The flowchart of the proposed underwater image enhancement algorithm.

Fig 4. Comparison of underwater image enhancement effect.

The upper line shows the original underwater images, and the lower line shows the enhanced images.

Fig 5. Covering effect of three-layer atrous convolution with equal atrous rate.

Fig 6. Covering effect of hybrid atrous convolution with unequal atrous rate.

Fig 7. The structure of the CASPP module.

Fig 8. The structure of the context information aggregation decoder.

Fig 9. The color representation of labeled categories.

Fig 10. Qualitative comparisons on colored cast underwater images.

From left to right are original images, the enhanced images, and the results generated by Deeplab V3+, DFANet, APCNet, the method proposed by Liu et al., our method, and the ground truth.

Fig 11. Qualitative comparisons on clear underwater images.

From left to right are the original images, the enhanced images, and the results generated by Deeplab V3+, DFANet, APCNet, the method proposed by Liu et al., our method, and the ground truth.

Fig 12. Failure examples of the proposed method.