Abstract: Camouflage nets in forested areas exhibit good concealment when subjected to single-band detection in natural environments. This study proposes a collaborative recognition method based on visible-light (380-1050 nm) and long-wave infrared (8-14 μm) hyperspectral data. A visible-light hyperspectral imaging system was used to obtain the color, shape, and spectral information of ground objects, whereas the long-wave infrared hyperspectral imaging system was employed to acquire bright-temperature spectral information. Multi-dimensional features such as spectral and textural features were extracted, and a random forest was used as the classification model to obtain the recognition results. The experimental results showed that the recognition effect of the dual-band fusion feature model was significantly better than that of the single-band model. The recall rate of the fusion model was 82.4%, which was 15.7% and 2.3% higher than those of the long-wave infrared and visible-light single bands, respectively. The F1 score increased by 11% and 2.16% compared to the long-wave infrared and visible light single bands, respectively. The dual-band fusion scheme effectively improved the recognition accuracy of the camouflage net, fully exploited the complementary advantages of dual-band spectral modes, and effectively solved the limitations of single-band detection.