Weight measurement plays an important role in shrimp grading process. In this paper, a machine vision based estimation of shrimp weight using contour area was proposed. A shrimp image capture system was designed which employs back lighting technology using a LED panel. A polynomial fitting was applied to calculate the relationship between the number of pixels inside the boundary contour of shirmp image and the weight of shrimp. The experimental results show that the weight estimation system archives high accuracy and it is feasible to use machine vision technology to measure the weight of shrimps. The proposed method is more accurate and faster than experienced shrimp grading workers.