**5. Conclusions**

In this paper, spectral and texture analysis of hyperspectral imaging technique were used to discriminate wheat powdery mildew without obvious symptoms and estimate corresponding disease severity. This study indicates that the combination of spectral and texture approaches is the sensitive feature for disease detection when building a PLS-LDA classification model and PLSR DS estimation model, which offered significantly improved accuracies in detection and quantification of wheat PM disease. The model based on selected sensitive features identified the disease even before the initiation of significant variations in the physiological and biochemical parameters of leaves after disease. Therefore, it can provide the basis for early detection, preventing and controlling the plant disease worldwide. The stable performance of the spectral and texture indices enabled the early detection of disease with more than 85% overall classification accuracies of PLS-LDA model, especially at early infection stage of 3–6 DAI with a DS of 1–6%. The DS can be well estimated by PLSR model with the *R*<sup>2</sup> value of 0.818 at booting growth stage. However, the present study is conducted on leaf scale with relatively few trials, and only used limited number of diseased and healthy leaves for classification on a daily basis after inoculation. In the future, we would evaluate the feature and model with a greater number of trials and data, and hope to expand to field and canopy level.

**Author Contributions:** Conceptualization, X.Y., H.L. (Hongyan Liu); methodology, H.L. (Haiyan Liu), H.L. (Hongyan Liu) and I.H.K.; software, H.L. (Hongyan Liu); validation, H.L. (Haiyan Liu); formal analysis, I.H.K.; investigation, I.H.K. and H.L. (Hongyan Liu); resources, W.L., X.W.; writing—original draft preparation, H.L. (Hongyan Liu) and H.L. (Haiyan Liu); writing—review and editing, H.L. (Haiyan Liu), X.Y., A.C., W.L., T.C., Y.T., Y.Z., W.C., supervision, X.Y.; project administration, X.Y.; funding acquisition, X.Y. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was supported by grants from the National Key Research and Development Plan of China (2019YFE011721), National Natural Science Foundation of China (31971780), the Key Projects (Advanced Technology) of Jiangsu Province (BE 2019383), 333 Project of Jiangsu Province (JS333), Collaborative Innovation Center for Modern Crop Production co-sponsored by Province and Ministry(CIC-MCP), the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD), and the 111 project (B16026).

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable. **Data Availability Statement:** Not applicable.

**Acknowledgments:** The authors would like to thank the reviewers for recommendations which improved the manuscript.

**Conflicts of Interest:** The authors declare no conflict of interest.
