Saliency Object Detection Research Based On Complex Domains

In the field of computer vision,saliency object detection is used in the image preprocessing stage,which can greatly reduce the amount of calculation and improve the accuracy and real-time performance of the detection system.Most of the currently proposed saliency target detection methods are solved on a single domain,due to environmental uncertainties and cluttered backgrounds in the surrounding area of the target,the significant target area for detection is inaccurate when the saliency map is generated,problems with unclear outlines and background misdetections often occur.Based on this,a saliency target detection method based on complex domain is proposed.First,multi-scale retinex enhancement algorithm is used to preliminarily process the original image in spatial domain.Then,undirected graph is established and node features on the images are extracted preliminarily.The hyper-complex Fourier transform in the frequency domain is reconstructed to acquire the smoothing amplitude spectrum,phase spectrum,and Euler spectrum.Then,background suppression graphs are obtained through the smoothness of multi-scale Gaussian kernel.At the same time,the multi-level feature of wavelet transform in the wavelet domain is utilized to extract multiple features in terms of images,and the saliency graph of multiple features is calculated.Finally,the proposed adaptive threshold selection method is used to fuse the background suppression diagram with the saliency graph of multiple features and the final saliency graph is obtained.Saliency object detection experiments in the standard test dataset images MSRA10 K,THUR15k and ECSSD are conducted,compared with the currently popular saliency target detection methods,the results show that under complex background and environmental disturbances,this method can accurately detect the significant target area and obtain a complete map with clear boundary contours.The evaluation index of the algorithm is significantly higher than that of the contrast algorithm,which can improve the accuracy while suppressing the cluttered background,have better robustness and universality.