Real-Time Detection Algorithm for Kiwifruit Canker Based on a Lightweight and Efficient Generative Adversarial Network.

Disease diagnosis and control play important roles in agriculture and crop protection. Traditional methods of identifying plant disease rely primarily on human vision and manual inspection, which are subjective, have low accuracy, and make it difficult to estimate the situation in real time. At present, an intelligent detection technology based on computer vision is becoming an increasingly important tool used to monitor and control crop disease. However, the use of this technology often requires the collection of a substantial amount of specialized data in advance. Due to the seasonality and uncertainty of many crop pathogeneses, as well as some rare diseases or rare species, such data requirements are difficult to meet, leading to difficulties in achieving high levels of detection accuracy. Here, we use kiwifruit trunk bacterial canker (Pseudomonas syringae pv. actinidiae) as an example and propose a high-precision detection method to address the issue mentioned above. We introduce a lightweight and efficient image generative model capable of generating realistic and diverse images of kiwifruit trunk disease and expanding the original dataset. We also utilize the YOLOv8 model to perform disease detection; this model demonstrates real-time detection capability, taking only 0.01 s per image. The specific contributions of this study are as follows: (1) a depth-wise separable convolution is utilized to replace part of ordinary convolutions and introduce noise to improve the diversity of the generated images; (2) we propose the GASLE module by embedding a GAM, adjust the importance of different channels, and reduce the loss of spatial information; (3) we use an AdaMod optimizer to increase the convergence of the network; and (4) we select a real-time YOLOv8 model to perform effect verification. The results of this experiment show that the Fréchet Inception Distance (FID) of the proposed generative model reaches 84.18, having a decrease of 41.23 compared to FastGAN and a decrease of 2.1 compared to ProjectedGAN. The mean Average Precision (mAP@0.5) on the YOLOv8 network reaches 87.17%, which is nearly 17% higher than that of the original algorithm. These results substantiate the effectiveness of our generative model, providing a robust strategy for image generation and disease detection in plant kingdoms.

Occurrence Regionalization of Kiwifruit Brown Spot in Sichuan.

Kiwifruit brown spot caused by Corynespora cassiicola is the most significant fungal disease in Sichuan, resulting in premature defoliation, which had a significant impact on yield and fruit quality. The objective of the study was to determine the occurrence regularity and suitability of kiwifruit brown spot in Sichuan. The occurrence of the disease in the main producing region was continuously monitored, the maximum entropy (MaxEnt) model was used to predict its potential distribution, and the key environmental variables were identified using the jackknife method. The results indicated that kiwifruit brown spot was widely distributed across the entire producing region in Sichuan, predominantly affecting the variety "Hongyang". The incidence (p < 0.01) and disease index (p < 0.05) showed a significant positive correlation with the cultivar, and decreased with the altitude increasing. The average area under the ROC curve (AUC) of 10 replicates was 0.933 ± 0.012, with an accuracy of 84.44% in a field test, confirming the reliability of the predicted results. The highly suitable distribution areas of kiwifruit brown spot were mainly located in the Chengdu and Ya'an regions. The entire Panzhihua region was an unsuitable distribution area, and the entire Pujiang County and Mingshan District were highly suitable distribution areas. The key environmental variables affecting the potential distribution of kiwifruit brown spot included isothermality (24.3-33.7%), minimum temperature in August (16.3-23.6 °C), maximum temperature in July (25.5-31.2 °C), minimum temperature in June (15.6-20.9 °C), precipitation in August (158-430 mm), and average temperature in October (15.6-18.8 °C). This study provides a theoretical basis for the reasonable layout of the cultivar and the precise prevention and control of the disease.

Identification and Genetic Characterization of Pseudomonas syringae pv. actinidiae from Kiwifruit in Sichuan, China.

Pseudomonas syringae pv. actinidiae causes kiwifruit bacterial canker and poses a major threat to the kiwifruit industry. This study aimed to investigate the genetic characteristics of the P. syringae pv. actinidiae population from kiwifruit in Sichuan, China. Sixty-seven isolates obtained from diseased plants were characterized using morphological features, multiplex-PCR, and multilocus sequence analysis (MLSA). The isolates exhibited the typical colony morphology of P. syringae pv. actinidiae. Multiplex PCR amplification identified every isolate as P. syringae pv. actinidiae biovar 3. MLSA of the three housekeeping genes gapA, gyrB, and pfk, revealed that the reference strains of the five described biovars were clearly distinguished by a combined phylogenetic tree, and all of the tested isolates clustered with the reference strains of P. syringae pv. actinidiae biovar 3. Through a phylogenetic tree constructed from a single gene, it was found that pkf gene alone could distinguish biovar 3 from the other biovars. Furthermore, all P. syringae pv. actinidiae isolates analyzed by BOX-A1R-based repetitive extragenic palindromic (BOX)-PCR and enterobacterial repetitive intergenic consensus (ERIC)-PCR clustered into four groups. The clustering results of BOX- and ERIC-PCR indicated that group III had the largest number of isolates, accounting for 56.72 and 61.19% of all 67 isolates, respectively, and the two characterization methods were similar and complementary. The results of this study revealed that the genomes of P. syringae pv. actinidiae isolates from Sichuan had rich genetic diversity but no obvious correlation was found between clustering and geographical region. This research provides novel methodologies for rapidly detecting kiwifruit bacterial canker pathogen and a molecular differentiation at genetic level of P. syringae pv. actinidiae biovar diversity in China.

Comparison and Correlation of Corynespora cassiicola Populations from Kiwifruit and Other Hosts Based on Morphology, Phylogeny, and Pathogenicity.

Corynespora leaf spot, which is caused by Corynespora cassiicola (Berk. & M. A. Curtis) C.T. Wei (C. cassiicola), has been globally reported in many plant species. 'Hongyang' was reported as highly sensitive kiwifruit cultivar to C. cassiicola. This cultivar is an important germplasm resource in the Actinidiaceae family and is widely cultivated throughout China. Even though C. cassiicola has been identified as the pathogen associated with kiwifruits in China, the C. cassiicola population from kiwifruit has not been characterized based on morphology, phylogeny, and pathogenicity. In this study, 133 and 48 representative C. cassiicola isolates from kiwifruit and 11 other hosts, respectively, recovered from symptomatic leaves were classified into eight morphological subgroups based on host origins. Using three loci (rDNA ITS, caa5, and act1), a phylogenetic tree showed that C. cassiicola isolates in Sichuan Province were grouped into three clades. All kiwifruit isolates were genetically identical to the rubber isolates from different countries. However, most isolates from other hosts in this study were genetically identical to the cucumber, soybean, and cowpea isolates in China, Brazil, and the United States, and two strawberry isolates clustered with isolates from tomato and other hosts in China, Brazil, and the United States. Furthermore, we confirmed host shift of C. cassiicola among different plant species in this study. Although 51 isolates from kiwifruit and different hosts were pathogenic to kiwifruit, blueberry, cucumber, and soybean, virulence levels of the pathogen were diverse for four hosts. Kiwifruit isolates exhibited host specificity with regards to the original host in degree. In addition, those isolates revealed a correlation between morphology and pathogenicity. The results suggest that C. cassiicola in Sichuan Province were derived from three different phylogenetic lineages. Promotion of the susceptible 'Hongyang' cultivar led to the emergence of a regnant C. cassiicola population from kiwifruit. In conclusion, rapid development of the C. cassiicola-sensitive crop in agricultural systems led to the emergence of a regnant C. cassiicola population. In some dominant populations (e.g., the C. cassiicola population from kiwifruit in this study), host origin was found to be a key factor influencing the morphologic, genetic, and pathogenic characterization of C. cassiicola.