Spectral sensitivities represent the spectral property of a digital camera. Most of the prior art spectral sensitivities estimation algorithms were applied to reflective colour charts, while some algorithms used a LED-based target. In this study, the spectral sensitivities of camera were estimated from both the LED-based and reflective colour targets. Four algorithms including Tikhonov Regularization based on Derivatives, Fourier basis function, Principal Component Analysis (PCA) and Singular Value Decomposition (SVD), were implemented. The estimated accuracy was compared between different types of colour targets, and between different algorithms. It was found that the optimal algorithm was different when using LED-based and reflective colour targets.