Rapid detection and classification of citrus fruits infestation by Bactrocera dorsalis (Hendel) based on electronic nose

A sweeping electronic nose system (SENS) was self-developed to detect the presence of early infestation by Bactrocera dorsalis (Hendel) in citrus fruits. Principal component analysis (PCA) and linear discriminate analysis (LDA) were applied to analyze citrus fruits that were subjected to different types of treatments (invasion and incubation stage) caused infestation. The results indicated that the SENS could successfully detect the presence of early infestation by B. dorsalis in citrus fruits. The different types of treatments in citrus fruits could be effectively classified by PCA and LDA, respectively. Meanwhile, the specific infestation time of citrus fruits within treatment stage could be satisfactorily identified by LDA model with correct recognition rate of 98.21%. Importantly, an optimized sensor array achieved better performance in classification and discrimination than that of the non-optimized. This study showed the potential feasibility of the electronic nose technology for in-filed detection of postharvest pest infestation citrus fruits under market conditions.