The detection of students’ behaviors in classroom can provide a guideline for assessing the effectiveness of classroom teaching. This study proposes a classroom behavior detection algorithm using an improved object detection model (i.e., YOLOv5). First, the feature pyramid structure (FPN+PAN) in the neck network of the original YOLOv5 model is combined with a weighted bidirectional feature pyramid network (BiFPN). They are subsequently processed with feature fusion of different scales of the object to mine the fine-grained features of different behaviors. Second, a spatial and channel convolutional attention mechanism (CBAM) is added between the neck network and the prediction network to make the model focus on the object information to improve the detection accuracy. Finally, the original non-maximum suppression is improved using the distance-based intersection ratio (DIoU) to improve the discrimination of occluded objects. A series of experiments were conducted on our new established dataset which includes four types of behaviors: listening, looking down, lying down, and standing. The results demonstrated that the algorithm proposed in this study can accurately detect various student behaviors, and the accuracy was higher than that of the YOLOv5 model. By comparing the effects of student behavior detection in different scenarios, the improved algorithm had an average accuracy of 89.8% and a recall of 90.4%, both of which were better than the compared detection algorithms.

The study of dynamic changes and spatial variation of landscape patterns is important to deeply understand the relationship between human activities and the natural environment. We selected a typical mountain area, Shizhu County, as the study area and analyzed the landscape’s dynamic changes and spatial variation in that area from 2000–2015. The results showed that cropland and forestland were the dominant landscape types in the study area. Cropland and grassland areas decreased, being mainly converted to forestland. Forestland and built-up land areas were increasing; the increase in built-up land was mainly due to the invasion into cropland areas, and the increase in forestland was mainly due to the conversion of cropland and grassland. Water bodies were affected by factors such as water storage in the Three Gorges Reservoir, and their area continued to increase. The change in landscape was most dramatic from 2005–2010, mainly due to the rapid increase in the areas of built-up land and water bodies and the rapid decrease in grassland area. There were apparent spatial variations in landscape distribution, patterns, and dynamic changes. Although water bodies were mainly distributed in the relatively gentle slope areas with an elevation of less than 200 m and a slope of 0°–6°, other landscapes were concentrated at an elevation higher than 500 m, a slope of 15°–35°, with a westerly or northwesterly aspect. These areas also had the most drastic landscape changes. At the type-level and the landscape-level, landscape indices showed greater variation with elevation and slope than with aspect. Finally, the variations with elevation, slope, and aspect differed among different landscape types.

No abstract

No abstract