Image-based phenotyping is a modern imaging technique that has emerged as a powerful tool for overcoming the limitations of traditional visual evaluation in chrysanthemum breeding. Visual scoring investigations can be labor-intensive, subjective, and inconsistent; therefore, breeding programs, which explore the number of cultivated traits and expand various trait complexities, that introduce an improved scoring method are needed. The advancement of digital imaging technologies has enabled the widespread use of digital cameras for collecting imaging data. Coupled with machine learning algorithms and cost-effective sensing technologies, digital imaging provides new opportunities for the accurate, rapid, and high-throughput evaluation of important traits. This mini-review summarizes the current image-based phenotyping techniques – including RGB imaging, computer-vision analysis, and deep learning using neural networks – and highlights their applications in investigating and evaluating flower color, morphology, and plant compact architecture, as well as stress responses to environmental and insect resistance. For floral breeders, special tools such as smartphonebased imaging and open-source analysis pipelines can be incorporated but their use is limited by their cost. Finally, after discussing the major challenges in chrysanthemum breeding, including those associated with various image capture techniques and the complexity of chrysanthemum floral forms, this review outlines future prospects for constructing image-derived traits with genomic data to predict the potential characteristics of trait developments. Furthermore, these developments emphasize the growing role of image-based phenotyping as an important complementary tool for improving the accuracy and efficiency of modern chrysanthemum breeding programs.