Development of composite material parts requires significant research and development effort. The fiber length, volume fraction and direction are important parameters in determining the properties of the part and are determined by extensive theoretical analysis, finite element analysis and experimental studies. Composite parts are now used in aircraft, space structures, and marine vessels extensively. Additive manufacturing (AM) methods are increasingly used for printing composite material parts because of their low production volume and complex structure, which is difficult to manufacture by other methods. Non-destructive imaging methods are widely used for quality assessment, for example micro-CT scan is used for porosity analysis in composite parts. Advancements in 3D scanning and imaging technology have raised a significant concern in reverse engineering of parts made by AM, which may result in counterfeiting and unauthorized production of high-quality parts. This work is focused on using imaging methods and machine learning (ML) to identify fiber direction in various sections of the composite for possible reverse engineering. Here not only the geometry is captured but also the tool path of 3D printing is captured in various sections of the microstructure. A dimensional accuracy with only 0.33% difference is achieved for the reverse engineered model.