Leaf area index (LAI), defined as half the total leaf area per unit ground surface area, and it is one of the most important parameters for modeling many forest ecosystem processes, e.g., canopy photosynthesis, evapotranspiration, respiration, and microclimates. Methods for directly measuring LAI mainly include destructive sampling, allometry, and litter collection. The LAI derived from these direct methods is close to the true LAI, which are often used to calibrate that LAI derived from optical methods. However, the former two methods are destructive, thus the litter collection method is the most useful direct method for estimating LAI in a forest stand. However, sorting leaf litter is time consuming because of substantial skill in identification, especially in a natural forest with high species diversity. Thus, the method for improving the efficiency of the litter collection method is needed urgently. The equal dominance model, stand dominance model and local dominance model could be used to accurately estimate the LAI based on the litter collection method and tree data (e.g., basal area BA; coordinate) have been verified by many previous studies in deciduous broadleaf forests. The equal dominance model assumes that leaf litter biomass of a litter trap can be partitioned equally to each species; The stand dominance model partitions leaf litter biomass to each species proportionally based on its relative dominance within a whole stand; The local dominance model partitions leaf litter biomass to species in proportion with their local relative dominance around each litter trap. However, few studies are reported to check whether these models are useful for estimating the LAI in a mixed needleleaf-broadleaf forest stand. In the present study, we estimated the LAI using the litter collection method in the mixed broadleaved-Korean pine forest, and these results as a reference are used to verify the effectiveness of these three models in predicting the LAI in the mixed needleleaf-broadleaf forest stand. In addition to, rely on Pinus koraiensis, Abies nephrolepis, Tilia amurensis, Acer mono, Betula costata and Ulmus laciniata, we explored the regression analysis between the LAI obtained by the litter collection method and basal area for each species. The results showed that the equal dominance model is not useful for predicting the LAI in a mixed needleleaf-broadleaf forest stand. The stand dominance model and local dominance model are useful for predicting the LAI, and the accuracies are over 86% and 90%, respectively. Furthermore, we found that at least 8 specific leaf areas (SLAs) of dominant species are to be obtained for accurate estimation of LAI in a mixed needleleaf-broadleaf forest stand. LAI correlated with BA significantly (P < 0.01) for the six species, with the smallest R² value of 0.67. These results lay a foundation for rapidly and accurately estimating the LAI in a mixed needleleaf-broadleaf forest stand in subsequent studies, and provide references for developing the regression analysis between the LAI and BA within non-destructive.