Today the leading cause of attack is traffic distribution that affects transmission as well as network performance. Traditionally the mechanism is utilized for detecting traffic distribution is not so accurate, user friendly and also they are time consuming. Therefore in this paper a hybrid approach is used that detect attacks by eliminating replicated features. It utilizes TDE with bisection mechanism that has great attack detection rate The result of the proposed system as compared to existing system is better by the margin of 8 to 10%.. The primary reason for the betterment is because of contrast enhancement and scaling factor at pre-processing and objective function overhead consideration at segmentation and classification phase. The results are shown in terms of accuracy, PSNR and MSE.