The imaging quality of the Mapping Imaging Spectrometer (IMS) is crucial for spectral identification and detection performance. In IMS, the image mapper significantly influences the imaging quality. Traditional image mappers utilize a single-point diamond machining process. This process leads to inevitable edge eating phenomena that further results in noticeable deficiencies in imaging, impacting spectral detection performance. Therefore, we propose a manufacturing process for the image mapper based on ultra-thin layered glass. This process involves precision polishing of ultra-thin glass with two-dimensional angles, systematically assembling it into an image mapper. The surface roughness after coating is generally superior to 10 nm, with a maximum angle deviation of less than 3'. This results in high mapping quality. Subsequently, a principle verification experimental system was established to conduct imaging tests on real targets. The reconstructed spectrum demonstrates excellent alignment with the results obtained from the Computed Tomography Imaging Spectrometer (CTIS). We thereby validate that this approach effectively resolves the issues associated with edge eating (caused by traditional single-point diamond machining), and leads to improved imaging quality. Also when compared to other techniques (like two-photon polymerization (2PP)), this process demonstrates notable advantages such as simplicity, efficiency, low processing costs, high fault tolerance, and stability, showcasing its potential for practical applications.