Fruit processing, packing, and agricultural production all depend heavily on the ability to identify apple harvesting time and conditions. Typically, farmers must endure a time-consuming and inefficient process of manual identification and prediction. In order to save human labor, expenses, and production time in the agricultural sector, a new categorization model that recognizes the best status, apple size and shape must be developed. Scientists may use the large dataset provided in this paper to create efficient algorithms that can identify a broader range of apple shapes and growth tracking and get around the constraints by improving efficiency and cutting down on calculation time. This dataset, which has a total sample size of 8,568 images, is made up of ten classes (Days 1–10) based on two collection times (morning and noon). Following that, the Metadata includes the crucial elements that affect Fuji growth, such as temperature, specific time, size, height, and weight. Data collection took place between October 1st and October 28th, 2023. China Agricultural University's College of Information and Electrical Engineering in Beijing, China, is the dataset's host.