Image classification technology and performance based on Deep Learning have already achieved high standards. Nevertheless, many efforts have conducted to improve the stability of classification via ensembling. However, the existing ensemble method has a limitation in that it requires extra effort including time consumption to find the weight for each model output. In this paper, we propose a simple but improved ensemble method, naming with Self-Weighted Ensemble (SWE), that places the weight of each model via its verification reliability. The proposed ensemble method, SWE, reduces overall efforts for constructing a classification system with varied classifiers. The performance using SWE is 0.033% higher than the conventional ensemble method. Also, the percent of performance superiority to the previous model is up to 73.333% (ratio of 8:22).