Click on “Download PDF” for the PDF version or on the title for the HTML version.
If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.
A Model to Determine Injection Position in Grass Carp Based on Machine Vision
Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan www.asabe.org
Citation: 2018 ASABE Annual International Meeting 1800151.(doi:10.13031/aim.201800151)
Authors: Dongdong Li, Jianping Li, Songming Zhu
Keywords: Grass carp; Vaccination; Injection position; Algorithm; Machine vision
Abstract. Vaccination has the advantages of low cost and high efficiency for building resistance against diseases in the grass carp, moreover it can reduce the use of antibiotics and improving food safety at the same time. Traditionally, grass carp vaccination has been performed manually, which is a laborious and time-consuming procedure with low accuracy. In order to resolve this issue, an automatic vaccination injection system with faster and more accurate procedure need to be developed. Determining the appropriate injection position, according to the size of fish, is the key to vaccinating successfully. In this study, grass carp, a popular freshwater cultured fish in China, was selected as the research object. The injection point in the abdominal region, which avoids internal organs, was determined by dissecting grass carp. The length(L), width(W) and the length between injection point and the end of the mouth of the fish(S) were measured using a scale. Through statistical analysis, the position of injection point can be expressed with L and W, S=0.436L-0.089(R² = 0.993). According to the expression, an image processing algorithm, which involves preprocessing, edge pattern matching, and edge point detection, was developed based on machine vision. The results showed that the injection error in the horizontal direction was 5.291%.
(Download PDF) (Export to EndNotes)