In situ ground truth data are an important requirement for producing accurate cropland type map, and this is precisely what is lacking at vast scales. Although volunteered geographic information (VGI) has been proven as a possible solution for in situ data acquisition, processing and extracting valuable information from millions of pictures remains challenging. This paper targets the detection of specific crop types from crowdsourced road view photos. A first large, public, multiclass road view crop photo dataset named iCrop was established for the development of crop type detection with deep learning. Five state-of-the-art deep convolutional neural networks including InceptionV4, DenseNet121, ResNet50, MobileNetV2, and ShuffleNetV2 were employed to compare the baseline performance. ResNet50 outperformed the others according to the overall accuracy (87.9%), and ShuffleNetV2 outperformed the others according to the efficiency (13 FPS). The decision fusion schemes major voting was used to further improve crop identification accuracy. The results clearly demonstrate the superior accuracy of the proposed decision fusion over the other non-fusion-based methods in crop type detection of imbalanced road view photos dataset. The voting method achieved higher mean accuracy (90.6–91.1%) and can be leveraged to classify crop type in crowdsourced road view photos.

The real-time parameter is a key point of protocol's performance evaluation. The most important thing is to give an effective and reasonable method to simulation. Based on the analysis of the Data Link layer of CAN bus protocol, a deterministic stochastic Petri nets model of MAC sub-layer in CAN bus protocol is built , aiming at the real-time performance analysis. Also a calculation formula is given combined with the actual example.