RESUMEN
To enhance food security, food safety, and environmental health, a bio-based integrated pest management (BIPM) strategy was evaluated at two coastal locations in Lebanon as an alternative to toxic pesticide sprays in commercial high-arched plastic tunnels common in many countries. The evaluation occurred during two cucumber and pepper cropping seasons: spring and fall. At each site, two commercial tunnels were used; farmers' conventional practices were applied in one tunnel, while the BIPM approach was followed in the second tunnel. In the farmers' practices, a total of 14 sprays of insecticide/acaricide mixtures were applied during the spring growing season, and 6 sprays were applied during the fall. In the BIPM tunnels, hotspot releases of local strains of Amblyseius swirskii and Phytoseiulus persimilis were applied. By the end of the spring season, the number of whitefly nymphs (WFNs)/leaf and thrips/leaf in the pesticide treatment were 4.8 and 0.06, respectively, compared to 0.1 and 0.33, respectively, in the BIPM treatment. Similarly, at the end of the fall season, the WFNs reached 19.7/leaf in the pesticide control as compared to 1.2/leaf in the BIPM treatment, proving the efficacy of A. swirskii. Farmers using conventional acaricides during both cropping seasons failed to control Tetranychus urticae, the two-spotted spider mite (TSSM). However, hotspot releases of P. persimilis were successful in controlling TSSM. By the end of June, the number of TSSMs reached 7.8/leaf in the BIPM treatment compared to 53/leaf in the pesticide treatment. Likewise, in December, TSSM numbers reached 9/leaf in the BIPM treatment compared to 40/leaf in the pesticide treatment. Preliminary observations of pepper showed that both predatory mites (A. swirskii and P. persimilis) gave similar or better efficacy against the three pests. The two local predatory phytoseiid mites seem to be effective in controlling these three major pests and to be adapted to local environmental conditions. A rate of increase of 0.86 was observed for P. persimilis and 0.22 for A. swirskii, in June, when maximum temperatures were close to 40 °C. This also shows a compatibility between the two predators. In conclusion, our BIPM approach was efficient under a Mediterranean climate in arched plastic tunnels with relatively poor aeration.
RESUMEN
Deep neural networks can be used to diagnose and detect plant diseases, helping to avoid the plant health-related crop production losses ranging from 20 to 50% annually. However, the data collection and annotation required to achieve high accuracies can be expensive and sometimes very difficult to obtain in specific use-cases. To this end, this work proposes a synthetic data generation pipeline based on generative adversarial networks (GANs), allowing users to artificially generate images to augment their small datasets through its web interface. The image-generation pipeline is tested on a home-collected dataset of whitefly pests, Bemisia tabaci, on different crop types. The data augmentation is shown to improve the performance of lightweight object detection models when the dataset size is increased from 140 to 560 images, seeing a jump in recall at 0.50 IoU from 54.4 to 93.2%, and an increase in the average IoU from 34.6 to 70.9%, without the use of GANs. When GANs are used to increase the number of source object masks and further diversify the dataset, there is an additional 1.4 and 2.6% increase in recall and average IoU, respectively. The authenticity of the generated data is also validated by human reviewers, who reviewed the GANs generated data and scored an average of 56% in distinguishing fake from real insects for low-resolutions sets, and 67% for high-resolution sets.
