Study On Relationship Between Appearance Quality And Texture Properties For Abalone

Abalone is a valuable seafood with high nutritional value and medicinal value.With the development of science and technology and market demand,abalone breeding technology is becoming more and more mature,and the scale and output of aquaculture are also increasing year by year.Abalone is made into various products or semi finished products for sale in the market.The rapid development of modern food industry has promoted the improvement of people’s living standards.The requirements for abalone products are not only limited to the content of nutrients and health conditions,but the texture characteristics of food has gradually been paid attention to by people.The texture characteristics of abalone are of great significance to the processing technology and detection of abalone products.At present,the commonly used method for determining the texture characteristics is TPA experiment,which requires special equipment,instruments and a series of late calculation.It can’t solve the standardization,industrialization and mass production of abalone products.Accordingly,a BP neural network model based on image processing technology is proposed to detect the texture characteristics of abalone quickly and nondestructively.The main contents are as follows:(1)The abalone was heat-treated in a constant temperature water bath at 60℃,80℃,85℃,90℃,and 100℃ for 0.5h,1h,2h,4h,and 6h,respectively.After abalone hot processing,TPA experiments were taken and industrial camera images were collected to obtain abalone pictures after heat treatment.(2)The abalone samples after heat treatment were tested by TPA.The specific parameters of elasticity,hardness,recovery and chewiness were calculated based on the texture parameters.(3)The apparent characteristics of abalone are extracted.The shape feature is to get the accurate three-dimensional model of abalone by three-dimensional reverse engineering,and extract the volume and surface area parameters under different heat treatment conditions,and get the change percentage of volume and surface area.Before the texture and color feature is extracted,the abalone image is preprocessed in order to remove redundant information and improve the image quality of abalone.The feature of texture is to extract four unrelated characteristic parameters in the image gray space using the MATLAB software in the gray space of abalone image,namely energy,contrast,correlation and entropy;The color features are in the RGB color space of the abalone image,using the color moments to extract 9 color features,reduce the dimension of the 9 color feature parameters,and finally extract three color principal components.(4)A model for the relationship between apparent characteristics and texture characteristics of abalone was established.The accuracy of the model prediction is 97.926%.(5)A software system for detecting the texture characteristics of abalone is designed.Integrated image processing technology and BP neural network algorithm are used to quickly predict abalone texture characteristics.