Abstract: The index of crop growth monitoring has a close relationship with crop yield. It could forecast a large-scale food state that indicates a possibility of either a missing or surplus yield as early as possible, and it is therefore important for the macro control of food. Using near-ground remote sensing is significant to understanding the growth of crops and providing accurate and scientific data for precision agriculture. For the small area growers, the vehicle-borne system shows the good prospects and has gradually become the first choice method. This paper discusses a method that is one of the most important tools for yield prediction for winter wheat in the jointing stage. It is an efficient, flexible, and economical operation for a small region. Usually the vehicle-borne growth monitoring system cannot maintain steady operations due to the row spacing of winter wheat in the jointing stage. The background interference on the reflectance will not be suppressed effectively, which will result in a deviation in the growth monitoring and yield prediction. In order to overcome this problem, a new vegetation index named MPRI (mobile photochemical reflectance index) was applied for the yield prediction in this paper. The MPRI derives from the PRI (photochemical reflectance index),which is defined as a normalized difference index using two narrow reflectance bands at 531 and 570 nm that are closely related to xanthophyll cycle pigment content. It has been successfully used to estimate leaf photosynthetic light use efficiency (LUE) across species which vary in water content and nitrogen concentration. Previous research studies demonstrated that a consistent relationship could be established between PRI and LUE calculated from gas exchange measurements at the leaf, small canopy, and full forestor crop canopy scales. It also showed some relationship between the PRI, LUE and wheat yield. The MPRI, which is proposed by this article, was constructed from the two reflectance bands which is a similar principle to PRI, and is constantly obtained by the vehicle-borne system sensors. The tests were carried out by the vehicle-borne system on the winter wheat field. The vehicle-borne system collected the reflectance data of the wheat canopy with the sensors at a sampling rate of 1 point per second. The GPS receiver obtained the location information at the same rate. The indexes of NDVI, TCARI, and MPRI were separately used for the diagnosis and analysis of the yield of wheat canopy, and finally their diagnosis results were contrasted. The results indicated that: It has satisfactory forecasting accuracy on the wheat yield by using the MPRI on the moving monitoring, and the R2 was about 0.76, which was same effect as much higher as by using NDVI. However, the MPRI has a better effect on removing the background interference. This is mainly because the canopy and the soil show the significant difference color. The MPRI of soil and winter wheat canopy are easily distinguishable by the threshold. By focusing on the yield spatial distribution, it was proposed that wheat yields, which were predicted by MPRI, were proven to be transformed with inverse distance weighted (IDW). It was proved that this method showed a positive effect on the yield prediction with the canopy reflectance in the jointing stage of wheat.