In-field automatic detection of maize tassels using computer vision

The heading stage of maize is an important period during its growth and development and indicates the beginning of its pollination. In this regard, an automated method for maize tassel detection is highly important to monitor maize growth. However, the recognition of maize heading stage mainly relies on visual evaluation. This method presents some limitations, such as expensive and subjective. This work proposed a novel method for automatic tassel detection. In the proposed algorithm, a color attenuation prior model was used to model the scene depth of saturation graph to remove image saturation. An Itti visual attention detection algorithm was used to detect the area of interest. Texture features and vegetation indices were used to develop a classification model to eliminate false positives. Pictures were captured using a commercial camera for two years to verify the stability of the proposed algorithm. Three indices were calculated to quantitatively assess and rate the algorithms. Experimental results show that the proposed method outperforms other existing methods, and its recall, precision, and F1 measure values are 86.30%, 91.44%, and 88.36%, respectively. Results indicate that the proposed method can effectively detect maize tassels in field images and remain stable with time.(c) 2020 China Agricultural University. Production and hosting by Elsevier B.V. on behalf of KeAi. This is an open access article under the CC BY-NC-ND license (http://creativecommons. org/licenses/by-nc-nd/4.0/).