Maosong Li

Bio: Maosong Li is an academic researcher. The author has contributed to research in topics: Convolutional neural network & Deep learning. The author has an hindex of 4, co-authored 5 publications receiving 54 citations.

Leaf Counting with Multi-Scale Convolutional Neural Network Features and Fisher Vector Coding

Abstract: The number of leaves in maize plant is one of the key traits describing its growth conditions. It is directly related to plant development and leaf counts also give insight into changing plant development stages. Compared with the traditional solutions which need excessive human interventions, the methods of computer vision and machine learning are more efficient. However, leaf counting with computer vision remains a challenging problem. More and more researchers are trying to improve accuracy. To this end, an automated, deep learning based approach for counting leaves in maize plants is developed in this paper. A Convolution Neural Network(CNN) is used to extract leaf features. The CNN model in this paper is inspired by Google Inception Net V3, which using multi-scale convolution kernels in one convolution layer. To compress feature maps generated from some middle layers in CNN, the Fisher Vector (FV) is used to reduce redundant information. Finally, these encoded feature maps are used to regress the leaf numbers by using Random Forests. To boost the related research, a relatively single maize image dataset (Different growth stage with 2845 samples, which 80% for train and 20% for test) is constructed by our team. The proposed algorithm in single maize data set achieves Mean Square Error (MSE) of 0.32.

Drought Stress Detection in the Middle Growth Stage Of Maize Based On Gabor Filter and Deep Learning

Abstract: Drought has become a major factor that limits maize production. This research is mainly focused on maize drought detection. The traditional detection method is mainly based on manual measurement. However it’s time consuming and costly, and some related methods may damaged to plants. In recent years, with the breakthrough of computer vision technology, measurement methods based on image processing technology have begun to be widely used. Image processing is not only low cost but also convenient for real-time analysis. According to some research, the water supply of maize in the two weeks before and after the pollination period will determine the final yield [1]. Therefore, the identification of drought in the middle of maize growth (we define the middle growth stage as 12-leaf to silking stage) is important for final yield. On the basis of this situation, an automatic detection system for drought stress in the middle growth stage of maize is proposed. We use different directions and wavelengths of Gabor filter to obtain the texture feature and then constitute a feature matrix after blocking and condensing features. Finally, the data were fed to the convolution neural network for secondary feature extraction and classification. The average recognition rate of the experiment is 98.84%. The final experimental results shows that our model has the adaptability to illumination and angle transformation, and it can also adapt to a complex field environment.

Identification and Classification of Maize Drought Stress Using Deep Convolutional Neural Network

Abstract: Drought stress seriously affects crop growth, development, and grain production. Existing machine learning methods have achieved great progress in drought stress detection and diagnosis. However, such methods are based on a hand-crafted feature extraction process, and the accuracy has much room to improve. In this paper, we propose the use of a deep convolutional neural network (DCNN) to identify and classify maize drought stress. Field drought stress experiments were conducted in 2014. The experiment was divided into three treatments: optimum moisture, light drought, and moderate drought stress. Maize images were obtained every two hours throughout the whole day by digital cameras. In order to compare the accuracy of DCNN, a comparative experiment was conducted using traditional machine learning on the same dataset. The experimental results demonstrated an impressive performance of the proposed method. For the total dataset, the accuracy of the identification and classification of drought stress was 98.14% and 95.95%, respectively. High accuracy was also achieved on the sub-datasets of the seedling and jointing stages. The identification and classification accuracy levels of the color images were higher than those of the gray images. Furthermore, the comparison experiments on the same dataset demonstrated that DCNN achieved a better performance than the traditional machine learning method (Gradient Boosting Decision Tree GBDT). Overall, our proposed deep learning-based approach is a very promising method for field maize drought identification and classification based on digital images.

Identifying crop water stress using deep learning models

Abstract: The identification of water stress is a major challenge for timely and effective irrigation to ensure global food security and sustainable agriculture. Several direct and indirect methods exist for identification of crop water stress, but they are time consuming, tedious and require highly sophisticated sensors or equipment. Image processing is one of the techniques which can help in the assessment of water stress directly. Machine learning techniques combined with image processing can aid in identifying water stress beyond the limitations of traditional image processing. Deep learning (DL) techniques have gained momentum recently for image classification and the convolutional neural network based on DL is being applied widely. In present study, comparative assessment of three DL models: AlexNet, GoogLeNet and Inception V3 are applied for identification of water stress in maize (Zea mays), okra (Abelmoschus esculentus) and soybean (Glycine max) crops. A total of 1200 digital images were acquired for each crop to form the input dataset for the deep learning models. Among the three models, performance of GoogLeNet was found to be superior with an accuracy of 98.3, 97.5 and 94.1% for maize, okra and soybean, respectively. The onset of convergence in GoogLeNet models commenced after 8 epochs with 22 (maize), 31 (okra) and 15 (soybean) iterations per epoch with error rate of less than 7.5%.