Tool wear plays a significant role for proper planning and control of machining parameters to maintain the product quality. However, existing tool wear monitoring methods using sensor signals still have limitations. Since the cutting tool operates directly on the workpiece during machining process, the machined surface provides valuable information about the cutting tool condition. Therefore, the objective of present study is to evaluate the tool wear based on the workpiece profile signature by using wavelet analysis. The effect of wavelet families, scale of wavelet and statistical features of the continuous wavelet coefficient on the tool wear is studied. The surface profile of workpiece was captured using a DSLR camera. Invariant moment method was applied to extract the surface profile up to sub-pixel accuracy. The extracted surface profile was analyzed by using continuous wavelet transform (CWT) written in MATLAB. The results showed that average, RMS and peak to valley of CWT coefficients at all scale increased with tool wear. Peak to valley at higher scale is more sensitive to tool wear. Haar was found to be more effective and significant to correlate with tool wear with highest R2 which is 0.9301.