With the prevalence of digital devices, images are now more accessible. A method to judge the image quality of a picture and corresponding datasets are highly desired. However, previous works focused solely on total image quality, without consider image quality separately in terms of color and spatial aspects. The present study aims to fill this gap by evaluating total, color, and spatial image quality together. The whole experiment was divided into two parts: no-reference (NR) experiment and with-reference (WR) experiment. In the NR part, 30 participants assessed total image quality (tIQ), color image quality (cIQ) and spatial image quality (sIQ) as well as their corresponding weights for color and spatial impact. In the WR part, 30 participants were asked to evaluate the difference in color and total image quality between the original image and rendered image. Weighted IQ, obtained through linear weighting using ratio, cIQ, and sIQ, demonstrated a high correlation coefficient (0.96) with total IQ. This implies that color and spatial features of image quality can be treated as separate entities. A no-reference image quality model was proposed to predict IQs whose accuracy of prediction obtained a correlation coefficient value of 0.80.