Memristor crossbar array (MCA) based accelerators can be used to accelerate neural networks. However, the conductances of memristors are slowly changed in the inference processes because of drift phenomena. As the weights are represented as the conductances, the weights are also changed slowly and the accuracy of neural network degrades. In order to slow down the accuracy degradation, some researchers propose methods using multiple memristors representing one weight, which leads to high area and energy overhead. Other researchers apply BFGS algorithm to determine the input amplitude and duration to reduce drift impact, which needs a huge amount of extra computation and is not practical. In this paper, we propose two techniques to slow down the accuracy degradation by recovering the accuracy after drifting. Firstly, we introduce one label memristor to monitor the drift degree of the crossbar and change the current-result conversion parameter to recover the accuracy. Secondly, we propose an auto-correction technique to correct the conductance of the label memristor. The conductance of the label memristor is used to change the parameter which is used in current-result conversion phase. Experimental results show that our proposed techniques can increase up to 8 times high accuracy (>95%) time and 3 times lifetime (>80%) for MCA based neural networks.