Abstract:The presented work is part of the H2020 project SWEEPER with the overall goal to develop a sweet pepper harvesting robot for use in greenhouses. As part of the solution, visual servoing is used to direct the manipulator towards the fruit. This requires accurate and stable fruit detection based on video images. To segment an image into background and foreground, thresholding techniques are commonly used. The varying illumination conditions in the unstructured greenhouse environment often cause shadows and overexposure. Furthermore, the color of the fruits to be harvested varies over the season. All this makes it sub-optimal to use fixed pre-selected thresholds. In this paper we suggest an adaptive image-dependent thresholding method. A variant of reinforcement learning (RL) is used with a reward function that computes the similarity between the segmented image and the labeled image to give feedback for action selection. The RL-based approach requires less computational resources than exhaustive search, which is used as a benchmark, and results in higher performance compared to a Lipschitzian based optimization approach. The proposed method also requires fewer labeled images compared to other methods. Several exploration-exploitation strategies are compared, and the results indicate that the Decaying Epsilon-Greedy algorithm gives highest performance for this task. The highest performance with the Epsilon-Greedy algorithm ( = 0.7) reached 87% of the performance achieved by exhaustive search, with 50% fewer iterations than the benchmark. The performance increased to 91.5% using Decaying Epsilon-Greedy algorithm, with 73% less number of iterations than the benchmark.

Root and butt-rot (RBR) has a significant impact on both the material and economic outcome of timber harvesting, and therewith on the individual forest owner and collectively on the forest and wood processing industries. An accurate recording of the presence of RBR during timber harvesting would enable a mapping of the location and extent of the problem, providing a basis for evaluating spread in a climate anticipated to enhance pathogenic growth in the future. Therefore, a system to automatically identify and detect the presence of RBR would constitute an important contribution to addressing the problem without increasing workload complexity for the machine operator. In this study, we developed and evaluated an approach based on RGB images to automatically detect tree stumps and classify them as to the absence or presence of rot. Furthermore, since knowledge of the extent of RBR is valuable in categorizing logs, we also classify stumps into three classes of infestation; rot = 0%, 0% < rot < 50% and rot ≥ 50%. In this work we used deep-learning approaches and conventional machine-learning algorithms for detection and classification tasks. The results showed that tree stumps were detected with precision rate of 95% and recall of 80%. Using only the correct output (TP) of the stump detector, stumps without and with RBR were correctly classified with accuracy of 83.5% and 77.5%, respectively. Classifying rot into three classes resulted in 79.4%, 72.4%, and 74.1% accuracy for stumps with rot = 0%, 0% < rot < 50%, and rot ≥ 50%, respectively. With some modifications, the developed algorithm could be used either during the harvesting operation to detect RBR regions on the tree stumps or as an RBR detector for post-harvest assessment of tree stumps and logs.

Abstract. It is essential to have a reliable system to detect humans in close range of forestry machines to stop cutting or carrying operations to prohibit any harm to humans. Due to the lighting conditions and high occlusion from the vegetation, human detection using RGB cameras is difficult. This paper introduces two human detection methods in forestry environments using a thermal camera; one shape-dependent and one shape-independent approach. Our segmentation algorithm estimates location of the human by extracting vertical and horizontal borders of regions of interest (ROIs). Based on segmentation results, features such as ratio of height to width and location of the hottest spot are extracted for the shape-dependent method. For the shape-independent method all extracted ROI are resized to the same size, then the pixel values (temperatures) are used as a set of features. The features from both methods are fed into different classifiers and the results are evaluated using side-accuracy and side-efficiency. The results show that by using shapeindependent features, based on three consecutive frames, we reach a precision rate of 80% and recall of 76%.

No abstract