Aquatic products are becoming increasingly popular because of their high nutritional value. Size information is an important parameter that can be used to measure the growth, weight, gender, grading and even species identification of fish. However, size information is a highly tedious and inefficient measure when conducted manually through traditional methods. Machine vision is a non-destructive, economic, rapid and efficient tool; hence, it is suitable to measure fish size. This review introduced methods and results for fish size measurement through machine vision. The paper is organised according to the measurement of body dimensionality: length measurement and area measurement. Simultaneously, the advantages, disadvantages and future trends of the system are discussed. With development in those areas, the size measurement by machine vision technology will become more effective. Machine vision system brings high accuracy and high efficiency and is easier than manual work. The methods reported can help researchers and farmers bring benefits for aquaculture.