The task of Camouflaged Object Detection (COD) is to identify and detect the camouflaged objects in images that are highly blended with the surroundings and difficult to distinguish. The camouflaged objects often have the similar colors and textures with their surroundings, and the visual cues in complex environments are quite weak, making COD tasks extremely challenging. Traditional methods of COD tasks are typically based on the single RGB input. In this paper, we introduce Hyperspectral Image (HSI) data into COD tasks, since HSI data provides rich spectral information which can be taken as the complement of RGB image features to break the camouflages. To effectively leverage the two modalities of RGB data and HSI data, we propose Hyperspectral Image-aided Camouflaged Object Detection method (HICOD), which consists of two modules: Multi-modal Feature Extraction (MFE) and Cross-modal Fusion (CF). MFE module extracts the features from RGB data and HSI data through a hybrid backbone network. CF module refines and fuses these features across modalities, thus enhancing and complementing the features for more accurate detection. Extensive experiments have been conducted on some COD datasets (CAMO, COD10K, and NC4K), and the experiment results demonstrate that our proposed HICOD significantly outperforms current state-of-the-art methods in terms of S-measure, weighted F-measure, E-measure, and mean absolute error.