Raman detection technology, with many advantages, has now become the preferred choice in the field of biochemical detection. Raman imaging is an important optionforRaman detection because it providesrich information of the concentration, distribution and changes of a certain substance component in the detection area. Raman spectrum for imaging construction obtained through experiments usually contains a limited number of pixels and the imaging effect is poor, so it is urgent to interpolate it. However, for most popular image interpolation methodscurrently available for conventional image processing, classical interpolation function is selected for interpolation based on the values and distribution of known pixel points. However, for Raman imaging, the existing interpolation method for conventional images is not enough for high quality imaging and the distribution of information at each pixel point (i. e., acquisition point) needs to betaken into account. Because the Raman signal collected at each pixel is from an objective lens and the light intensity is Gaussian distribution, the relationship ofthe collected Raman scattering signal at each pixel and the spatial distribution is also Gaussian dependent. That is, the signal center (i. e., the acquisition center point) has the highest signal ratio, and the signal around the collection point is Gaussian dependent. This rule tells us that the Raman scattering signal collected at the collection point actually contains the signal of the surrounding points. Based on the special significance of Gaussian beam in Raman imaging, this paper theoretically analyzes and proposes a new interpolation method suitable for Raman imaging. After setting a specific acquisition interval to make sure that the acquisition area at the adjacent collection point is tangent, we use our function based on the information contained at the collection point for interpolation, which meets the interpolation requirement. Unlike the conventional interpolation method, we establish an appropriate relationship between the Gaussian beam and the acquired Raman signal so that the interpolation can also indirectly reflect the biological information of the collected area. By comparing the imaging before and after interpolation, it is found that the new interpolation method has a good amplification effect, and the amplification by this interpolation method can greatly save the acquisition time for imaging construction with similar resolution ratio and information, there by saving experimental resources. We take the interpolation and amplification twice as an example, and explain the interpolation method in detail. In reality, the corresponding interpolation amplification can be performed according to the specific requirements.