Grape leaves detection and tracking based on improved deformable part model and discriminative model

Recently, some researchers have exploited computer vision based video analysis to monitor the growth status of crop under the natural condition. Since leaf is the largest organ of the vast majority of plants, it often serves as primary monitoring object. Most of analysis algorithms of illness detect the blobs on the leaf surface and then judge the kind of diseases. In a leaf image, the blobs may be caused by shadow, dust, highlight, and so on, which are prone to be confused with the blobs caused by diseases. To accurately judge the illness of a leaf for online surveillance, it is important to consider the time factor, since the blobs caused by the aforementioned factors may disappear with time elapsing. There exist some reasons such as the various poses, mutual occlusion, appearance and the irregular movement, which make the conventional detection and tracking methods hard to locate the leaves accurately in the images. In this paper, a novel scheme to monitor the leaves of vine grape was proposed. To improve the accuracy of leaves detection, a traditional RGB (red, green, blue) image was replaced by a G/R image to train the deformable part model (DPM) since the former makes it easier to distinguish the grape leaves from the background than the latter. The DPM detector for leaves was constructed based on HOG feature, which was a mixture over 3 components representing different aspects of a leaf. Since high dimension of HOG feature hampered real-time detection, PCA (principal component analysis) method was exploited to reduce its dimension, which speeded up the process of training and detection effectively. By utilizing the trained model, the overall score was computed for each root location according to the best possible placement of the parts through the matching procedure. The scores were sorted and the detection with the highest score was picked out. To robustly trace the sharp movement of a leaf, probability model based online object tracking algorithm with color features was put forward. In the proposed algorithm, object-background model capable of differentiating a leaf from the background was constructed firstly. To reduce the risk of drifting towards the regions which exhibit similar appearance of leaf (but not real leaf) at a next frame, then a distractor-aware representation was combined to the formal model to generate a discriminative object model. Based on this model, detection rate and false detection rate were computed. This allows us to efficiently obtain the new object location in the next frame. In the long-term tracking process, detection repeated at the 30-minute interval to check whether new leaves appeared in the vision field or not. For the sake of the robustness, the images were gathered at various conditions, such as sunny day, cloudy day, shadow, flowering stages and fruiting stages to train the detection and tracking models. Experiments were conducted to evaluate the performance of leaf detection at 5 different settings. The experimental results showed that the average detection rate reached up to 88.31%, and the average false detection rate fell down to 8.73%. For the tracking algorithm, the results were also exciting: The overlap rate was as high as 0.83, and the average center error was 17.33 pixels. Compared with the classical KLT (Kanade-Lucas-Tomasi) tracking algorithm, our algorithm demonstrated the better robustness in the condition of illumination change and sharp movement. © 2017, Editorial Department of the Transactions of the Chinese Society of Agricultural Engineering. All right reserved.