In order to realize the real-time detection of tobacco leaf βcarotene content during curing, six color characteristic parameters of tobacco leaf obtained by image processing were used for quantification of the changes of color feature, and the quantified six color characteristic parameters were used as the input to the BP neural network, with the output being the content of βcarotene. The test result showed the red，green，blue color components and brightness of tobacco leaves increased first, and then decreased during curing, the hue of tobacco leaves decreased significantly during curing, and the saturation of tobacco leaves changed slowly. The correlation coefficient of the prediction model was 0.982, and the average relative error was 0.092, which met the requirements of actual bulk curing and provided a theoretical reference for the application of image processing technology during the bulk curing.