Camellia Recognition Image Dataset for Garden Flowers

Currently, garden flower recognition plays an important role in smart agriculture and horticulture management. However, with many camellia varieties that appear similar, recognition becomes difficult. Existing solutions often rely on a small number of samples or manual experience, which limits accuracy. This dataset aims to improve recognition algorithm accuracy with extensive camellia image resources, meeting the needs of automated recognition business requirements. Data collection is conducted with high-resolution cameras, covering various camellia varieties and different blooming stages to ensure sample diversity and representativeness. Quality control involves multiple rounds of annotation and expert review, with an annotation team consisting of botanists and image recognition experts. The data format is JPG, structurally stored with detailed annotation files. Data preprocessing involves image enhancement and normalization steps to enhance training effectiveness.