A machine vision system is developed for measurement and comparison of biological shapes such as the mouse vertebrae. The system is flexible and able to work under varying illumination con- ditions. The rate of growth and shape change of the vertebrae are evaluated quantitatively by using a new pattern recognition technique. The image segmentation process is made difficult since these im- ages are plagued by poor contrast and dropouts. In this paper, a review of previous work is presented, along with how this problem can be viewed in the context of the computer vision area. The system consisting of a a video camera (Panasonic CCTV), digitizing unit with framestore, optics and a microcomputer measures the dimensions and compares the shapes of complex biological structures. The image processing system helps automating the measurement problem of such complex shapes and objectifies the measurement results. Reproducibility is an interesting feature of the developed system. An assessment of the measurement accuracy and time duration was undertaken. Different steps in the implementation of this solution are discussed and results are presented. Although our ultimate goal is automatic measurement of biological shape, attention will be restricted to a fast method for both parallel outlining of the vertebrae and feature extraction. Experimen- tal results on mouse vertebrae are presented to successfully demonstrate the feasibility of the method for low quality images.