Abstract
The target detection model based on convolutional neural networks has recently achieved a series of exciting results in the target detection tasks of the PASCAL VOC and MS COCO data sets. However, limited by the data set for a particular scenario, some techniques or models applied to the actual environment are often not satisfactory. Based on cluster analysis and deep neural network, this paper proposed a new Statistic Experience-based Adaptive One-shot Network (SENet). The whole model solved the following practical problems. (1) By clustering the existing image classification dataset ImageNet, a common set of target detection datasets is formed, and a data set named ImageNet iLOC is formed to solve the object detection. The problem of single and insufficient quantities in the task. (2) We use cluster analysis on the size and shape of objects in each sample, which solves the problem of inaccurate manual selection of suggested areas during object detection. (3) In the multi-resolution training and prediction process, we reasonably allocate the size and shape of the suggested frame at different resolutions, greatly improve the utilization rate of the proposed frame, reduce the calculation amount of the model, and further improve the real-time performance of the model. The experimental results show that the model has a breakthrough in accuracy and speed (FPS reaches 54 in the case of a 3.4% increase in mAP).
Highlights
Convolutional neural network (CNN) was widely used in the 1990s, but with the rise of support vector machines in the field of computer vision, CNN entered a period of low tide
Target detection training set and test set are Passcal VOC 2007, 2012, COCO, and our ImageNet iLOC data set based on Section 3.1
3.1 Experimental results in PASCAL VOC 2007 On this data set, our SENET method is compared with SSD, YOLO, and Faster Region proposal based CNN (R-CNN)
Summary
Convolutional neural network (CNN) was widely used in the 1990s (such as model [1]), but with the rise of support vector machines in the field of computer vision, CNN entered a period of low tide. The above work has achieved good results in the identification of plant diseases, the challenges of complex field conditions, infection changes, various pathologies in the same image, and surrounding objects have not been studied They mainly use images acquired in the lab, so they cannot handle all the situations that occur in real scenes. Fuentes et al [26] proposes a system that can successfully detect and locate nine Cole pests and diseases using images collected in the field, including actual cultivation conditions This approach differs from other methods in that it generates a set of bounding boxes that contain the location, size, and category of the disease and/or pest in the image. Our method has studied several techniques to make the system more robust to inter- and intra-species changes in Cole pests and diseases
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: EURASIP Journal on Wireless Communications and Networking
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
