When a typhoon or natural disaster occurs, a significant number of orchard fruits fall. This has a great impact on the income of farmers. In this paper, introduced an AI-based method to enhance low-quality raw images. Specifically, it focus on apple images, which are being used as AI training data. This model utilize both a basic program and an artificial intelligence model to conduct a general image process that determines the number of apples in an apple tree image. The main objective is to evaluate high and low performance based on the close proximity of the result to the actual number. The artificial intelligence models utilized in this study include the inception-V3, Convolutional Neural Network (CNN), VGG16, and Random Forest models, as well as a model utilizing traditional image processing techniques. The study found that 53 red apple fruits out of a total of 87 were identified in the apple tree image, resulting in a 62% hit rate after the general image process. The VGG16 model identified 66, corresponding to 90%, while the Random Forest model identified 37, corresponding to 85.4%. The CNN model identified 51, resulting in a 97.6% confirmation rate and inception-V3 model identified 52, resulting in a 99.2%. Therefore, selected an artificial intelligence model with outstanding performance and use a real-time object separation method employing artificial function and image processing techniques to identify orchard fruits. This application can notably enhance the income and convenience of orchard farmers.