VegeNet - Image datasets and Codes

Compilation of python codes for data preprocessing and VegeNet building, as well as image datasets (zip files). Image datasets: <strong>vege_original</strong> : Images of vegetables captured manually in data acquisition stage <strong>vege_cropped_renamed</strong> : Images in (1) cropped to remove background areas and image labels renamed <strong>non-vege images</strong> : Images of non-vegetable foods for CNN network to recognize other-than-vegetable foods <strong>food_image_dataset</strong> : Complete set of vege (2) and non-vege (3) images for architecture building. <strong>food_image_dataset_split</strong> : Image dataset (4) split into train and test sets <strong>process</strong> : Images created when cropping (pre-processing step) to create dataset (2).本数据集包含用于数据预处理与VegeNet构建的Python代码集，以及压缩包格式的图像数据集。各图像数据集详情如下：<strong>vege_original</strong>：数据采集阶段手动拍摄的蔬菜图像；<strong>vege_cropped_renamed</strong>：对<strong>vege_original</strong>中的图像进行背景抠除与图像标签重命名操作后得到的图像；<strong>non-vege images</strong>：用于卷积神经网络（Convolutional Neural Network, CNN）识别非蔬菜类食物的非蔬菜食物图像；<strong>food_image_dataset</strong>：涵盖<strong>vege_cropped_renamed</strong>与<strong>non-vege images</strong>全部图像的完整数据集，用于模型架构构建；<strong>food_image_dataset_split</strong>：被划分为训练集与测试集的<strong>food_image_dataset</strong>；<strong>process</strong>：创建<strong>vege_cropped_renamed</strong>数据集时，通过抠图预处理步骤生成的中间图像。