Decapterus Macarellus Rott and Fresh Model

Pelagic fish such as mackerel are a source of protein in Indonesia. However, there is no decapterus macarellus as an open dataset for image processing using various classification algorithms. Where its use includes the sensor-assisted sorting process in checking fresh fish and rotten fish. For this reason, this study aims to provide a classification model for pelagic fish and their primary datasets which is available for free on the IEEE data port. Artificial intelligence is used in the process of guided classification with the help of ground truth for the preparation of fish classes. The dataset used is a primary dataset consisting of fish images, arranged in two classes, namely 71 fresh fish and 96 rotten fish. The methods used are k-NN classifiers, naive bayes and ridge regression. Experiments were drawn up to classify rotten fish and fresh fish. The preprocessing was assisted by InceptionV3 as a feature extraction method. Furthermore, the image data is trained in a ratio of 60:40 for training and testing data. Validation was performed using 2-fold cross validation with the results obtained being 99.4%, 94% and 100% accuracy using the classifiers namely k-NN, naive bayes and ridge regression, respectively.在印度尼西亚，鲭鱼等远洋鱼类（pelagic fish）是重要的蛋白质来源。然而，目前尚无针对圆鲹（Decapterus macarellus）的开源数据集，以供各类分类算法开展图像处理研究，此类数据集可应用于传感器辅助的鲜鱼与腐鱼分拣检测流程。为此，本研究旨在构建远洋鱼类分类模型并配套提供原始数据集，该数据集可于IEEE数据端口（IEEE Data Port）免费获取。本研究借助真值标注（Ground Truth）完成鱼类类别的标注筹备工作，并利用人工智能开展引导式分类流程。本研究使用的原始数据集包含两类鱼类图像：71张鲜鱼图像与96张腐鱼图像。研究所采用的分类方法包括k近邻（k-NN）分类器、朴素贝叶斯（Naive Bayes）与岭回归（Ridge Regression），实验旨在实现鲜鱼与腐鱼的分类任务。预处理阶段采用InceptionV3作为特征提取（Feature Extraction）方法。随后，图像数据以60:40的比例划分为训练集与测试集，并采用2折交叉验证（2-fold Cross Validation）开展模型验证。最终，k近邻、朴素贝叶斯与岭回归分类器的分类准确率分别达到99.4%、94%与100%。

搜集汇总

背景与挑战

背景概述

该数据集是一个用于鲭鱼（Decapterus macarellus）新鲜与腐烂状态分类的机器学习数据集，包含167个样本（71个新鲜鱼和96个腐烂鱼），每个样本有2048个基于InceptionV3提取的图像特征。数据集旨在支持传感器辅助的鱼类分选过程，并已通过k-NN、朴素贝叶斯和岭回归等分类算法验证，最高准确率达100%。