Hyperspectral imaging(HSI)technology，which can simultaneously capture spatial and spectral information， has substantial application potential in fields such as remote sensing， biomedicine，and material analysis. However，conventional mechanical scanning systems suffer from low efficiency， limited light throughput， and insufficient stability， restricting their widespread practical deployment. In recent years，digital micromirror devices(DMD)have been widely adopted to optimize and innovate hyperspectral imaging systems，providing high-speed modulation，broad spectral response，and excellent reliability. This review provides a systematic overview of DMD-based HSI， covering push-broom， coded-aperture， snapshot， microscopic， video-based， and ultrafast transient imaging architectures， and presents a unified quantitative comparison of their performance. The research results show that the spectral coverage of DMD systems has been extended from the visible range to near-infrared windows；spectral resolution ranges from 3. 73 nm in multiplexed systems to 0. 2 nm in tunable architectures；imaging speed includes 10 Hz spectral video and >30 fps hyperspectral video，and even reaches a frame rate of 3. 85 THz in ultrafast systems. In addition，this review examines the engineering challenges of the system，such as stability，alignment tolerance， environmental robustness， and computational burden， and outlines future research directions such as miniaturization，intelligent control，and multimodal integration.