The multispectral imaging technique has been used for skin analysis, especially for distant mapping of in-vivo two major skin chromophores: melanin and blood. Previously, we have developed a new LED-based MSI system. Our MSI system captures only 11 monoband images of human skin which is too little for providing an accurate diagnostic information. We implemented artificial neural network (ANN) algorithm to retrieve a hyperspectral reflectance cube between 380 and 880 nm with a 5 nm resolution. In this paper, we present the validation results our MSI's performance in order to estimate melanin and blood content in human skin. Validation is done by comparing the estimation results captured by the MSI system and the spectrometer with diffuse reflectance spectroscopy (DRS) technique. For validation, we used agar-based phantom as object. In addition, we also present the results of in vivo testing in uper human palm skin with 5 samples. The estimation results of melanin content by using MSI system and the spectrometer follows the equation y = 0.9565x with the coefficient of determination (R2) of 0.9934. On the other hand, the estimation result of the blood content follows the equation y = 1.1881x with the coefficient of determination (R2) of 0.9226. The average melanin content estimate results with the MSI system was 1.29% greater than the spectrometer, while the average blood content was 1.06% higher for the same in vivo skin data test. According the results, the LED-based MSI can be used for the direct and noninvasive skin assessment accurately. For example, measurement of melanin content in the skin is the most essential in the skin diagnosis process, especially in cases of hypo-pigmentation and hyper-pigmentation. In other cases, some cancers such as melanoma, basal cell carcinoma (BCC), and the tumor can also be differentiated by measurement of melanin and blood content.