The Measurement of Fish Size by Machine Vision - A Review
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 i
- PDF / 2,481,709 Bytes
- 18 Pages / 439.37 x 666.14 pts Page_size
- 99 Downloads / 196 Views
1
)
College of Information Technology, Jilin Agricultural University, Changchun 130118, China [email protected], [email protected] 2 College of Information and Electrical Engineering, China Agricultural University, Beijing 100083, China [email protected]
Abstract. 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 nondestructive, 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. Simultane‐ ously, 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. Keywords: Machine vision · Fish size · Fish length · Fish area
1
Introduction
Fish and fish products play an important role in meeting human protein demand. To sustainably manage fish resources with high efficiency, fishers have to catch fish appro‐ priately and obtain meaningful information [1]. One important piece of information is fish size [2], which is a key parameter to appraise stock status, provide management advice and enhance economic benefits for an aquaculture enterprise [3]. In fish culture, gathering fish size information to describe and manage the growth of fish is necessary to harvest the stock at the optimum time. Furthermore, length distribution of the fish can be used to predict a range of population-level characteristics [4, 5] and to assess the stock scientifically during rearing [6–8]. Whether for aquaculture or market purposes, fish grading is an important and frequent operation [9]. During rearing, fish need to be graded by size; fishers can separate the fast growers from the slow ones to adjust the distribution of food and management to reduce cost [10–12]. For fish farms to adapt to the market, fish should be sorted by size after harvesting to estimate the value. This step © IFIP International Federation for Information Processing 2016 Published by Springer International Publishing AG 2016. All Rights Reserved D. Li and Z. Li (Eds.): CCTA 2015, Part II, IFIP AICT 479, pp. 15–32, 2016. DOI: 10.1007/978-3-319-48354-2_2
16
M. Hao et al.
is necessary to set different prices and manage markets conveniently [13–17]. Simulta‐ neously, size information can be used to monitor quality in the food industry, which can help fishers assess the price. When many kinds of fish are reared in one pond or net,
Data Loading...
