Abstract: Based on the digital image processing technique,a method was presented for microfabric analysis for ideal granular materials composed of metal bars with specified sections.A code named IPFA was then developed to implement the functions such as particle detection,contact searching and microfabric analysis.By use of this method,the digital images of dense samples were firstly enhanced to clear the noises.The particles were then detected from the enhanced images using template matching algorithm presented here,and contact normals between particles were searched based on the particle detection results.Finally,the statistical analysis could be processed on two important elements of microscopic fabric,including directions of the long axis of particles and the contact normals between particles.The proposed method was especially suitable for ideal two-dimensional granular materials,and could be used as a powerful tool for the analysis of microfabric of granular materials and its evaluation because of its high precision and efficiency of particle detection.