It is important to detect floating oil slicks after spill accidents, and hyperspectral remote sensing technology is capable of achieving this task. Traditional methods mainly utilize the spectral indices of hydrocarbons to detect floating oil slicks, but are poor at distinguishing the thickness of oil slicks and cannot detect sheens. Since the spectra of oil slicks should be affected by seawater as well as oil, this paper investigated the use of spectral indices of hydrocarbons and seawater to identify different thicknesses of oil slicks. In this research, a measurement, called index separability (IS), was proposed for quantitatively evaluating the identification ability of these spectral indices. Based on the evaluation results, experiments were conducted to validate the applicability of these spectral indices. The results show that the spectral indices of hydrocarbons are more suitable for detecting continuous true color oil slicks and emulsions and that spectral indices of seawater are more suitable for sheens and seawater. In addition, the spectral indices of hydrocarbons and seawater are complementary for detecting oil slicks. Finally, combining the spectral indices of hydrocarbons and seawater is conducive to achieving more accurate oil slick recognition results.

Oil spill incidents threaten the marine ecological environment. Detecting sea surface oil slicks by remote sensing images provides support for the efficient treatment of oil spills. This is important for sustainable marine development. However, traditional methods based on field analysis are time-consuming. Spectral indices lack applicability. In addition, traditional machine learning methods strictly rely on training and testing samples which are in short supply in oil spill images. Inspired by the spectral DNA encoding method, a spectral gene extraction (SGE) method was proposed to detect oil spills in hyperspectral images (HSI) and multispectral images (MSI). The SGE method contained a parameter and two strategies. The parameter of elimination was designed based on the population genetic frequency. It was used to control the number of spectral genes. The spectral gene extraction strategies, named largest in-class similarity (LIS) strategy and largest inter-class difference (LID) strategy, were proposed to mine the spectral genes by oil spill samples. The oil spills would be determined by calculating the similarity of the extracted spectral genes to the DNA encoded images. In this research, the SGE method was validated by two AVIRIS images of the Gulf of Mexico oil spill, one MODIS image of the Gulf of Mexico oil spill, and one Landsat 8 image of a Persian Gulf oil spill. The oil spills in different remote sensing images could be detected accurately by the proposed method in a small set of samples. Experimental results indicated that the proposed method was suitable for detecting marine oil spills in AVIRIS, MODIS, and Landsat 8 images. In addition, the SGE method with the LIS strategy was more suitable for detecting oil spills in HSI. Its proper elimination rates were 0.8~1.0. The SGE method with the LID strategy was more suitable for detecting oil spills in MSI. Its proper elimination rates were 0.5~0.7.

To improve the accuracy of change detection and solve the issues of missed and false detections in change detection, we conducted research from two perspectives: difference image saliency detection and object-based change vector analysis (CVA). A saliency-guided change detection algorithm was proposed and the CVA method was extended to an object-oriented level. First, we calculated the difference images of the corresponding band for the bi-temporal image and then performed principal component analysis to extract its first component for convolution, filtering, and other processing to obtain the saliency region. Furthermore, bi-temporal images were segmented and split to obtain sub-objects, extract various features, such as the spectrum, texture, and shape, and calculate their differences. The main features were optimized using VarSelRF; the initial change region was then obtained using object-based CVA for change detection. The final change region was obtained by combining the assumption that the saliency and intensity of the change region were not uniformly weak with fuzzy rules. This study selected 6 different complex images for experiments. The results demonstrated that the proposed saliency algorithm for difference images could effectively identify change regions, showing significant advantages compared to classical saliency algorithms. The proposed saliency-guided change detection algorithm achieved a high extraction accuracy, with an overall accuracy and Kappa coefficient typically exceeding 92% and 0.82, respectively. Even in complex environments, the Kappa coefficient reached 0.71. A comparison of the saliency-guided change detection with the pixel-level CVA and ablation experiments using two individual algorithms extraction demonstrated the superiority of the proposed method.