The detection of double compression plays an important role in JPEG image forensics and steganalysis. This paper introduces a new statistic called factor histogram, which describes the distribution of the factors being related to the quantized discrete cosine transform coefficients of JPEG image. Then, two concrete schemes based on factor histogram are presented for detecting doubly compressed images and identifying primary quality factor, respectively. Our experimental results demonstrate that both of the proposed schemes perform well, which validates the applicability of factor histogram in the detection of double compression and the estimation of primary quantization parameter.