US 11,900,584 B1
Method for judging freshness of cultured fish product based on eye image recognition
Lanlan Zhu, Shandong (CN); Xudong Wu, Shandong (CN); Xiuting Wei, Shandong (CN); Qingxiang Zhang, Shandong (CN); Hengjia Ni, Shandong (CN); Ruining Kang, Shandong (CN); and Lei Liu, Shandong (CN)
Assigned to SHANDONG UNIVERSITY OF TECHNOLOGY, Shandong (CN)
Filed by SHANDONG UNIVERSITY OF TECHNOLOGY, Shandong (CN)
Filed on Aug. 31, 2023, as Appl. No. 18/459,385.
Claims priority of application No. 202211451493.1 (CN), filed on Nov. 21, 2022.
Int. Cl. G06T 7/00 (2017.01); G06T 7/11 (2017.01); G06T 7/136 (2017.01); G06V 40/18 (2022.01); G06V 20/68 (2022.01); G06T 7/12 (2017.01)
CPC G06T 7/0004 (2013.01) [G06T 7/11 (2017.01); G06T 7/12 (2017.01); G06T 7/136 (2017.01); G06V 20/68 (2022.01); G06V 40/18 (2022.01); G06T 2207/30128 (2013.01)] 5 Claims
OG exemplary drawing
 
1. A method for judging freshness of cultured fish product based on eye image recognition, comprising following steps:
obtaining an eye area and an eye center point of the cultured fish product according to a fish head area of the cultured fish product;
obtaining each gray scale change sequence of the eye area according to coordinates of each pixel in each direction of the eye area and a gray scale value of the each pixel; obtaining a first data category and a second data category of the each gray scale change sequence according to the gray scale value of the each pixel in the each gray scale change sequence; recording an average gray scale value of the each pixel in the first data category as a first mean value, and recording an average gray scale value of the each pixel in the second data category as a second mean value; taking a difference value between the second mean value and the first mean value of the each gray scale change sequence as an overall gray scale difference of the each gray scale change sequence; obtaining a fish eye turbidity of the eye area according to the first data category, the second data category and the overall gray scale difference of the each gray scale change sequence; and
obtaining a gray scale difference of the each pixel in the first data category according to the gray scale value of the each pixel and the first mean value in the first data category of the each gray scale change sequence; obtaining a first non-plumpness of the eye area according to a distance from the each pixel in the first data category to the eye center point and the gray scale difference of the each pixel; obtaining a second non-plumpness of the eye area according to the second data category of the each gray scale change sequence; obtaining a fish eye plumpness of the eye area according to the first non-plumpness and the second non-plumpness of the eye area; and
obtaining the freshness of the cultured fish product according to the fish eye turbidity and the fish eye plumpness in the eye area;
wherein, a method for obtaining the first data category and the second data category of the each gray scale change sequence is:
obtaining a segmentation threshold of the each gray scale change sequence, and recording a maximum distance from the each pixel with the gray scale value smaller than the segmentation threshold in the each gray scale change sequence to the eye center point as a boundary distance; dividing the pixel with a distance from the eye center point less than or equal to the boundary distance in the each gray scale change sequence into the first data category; dividing the pixel with the distance from the eye center point greater than the boundary distance into the second data category; and sequentially processing the each pixel in the each gray scale change sequence to obtain the first data category and the second data category of the each gray scale change sequence.