Improved YOLO v5 with balanced feature pyramid and attention module for traffic sign detection

Abstract

With the development of automatic driving technology, traffic sign detection has become a very important task. However, it is a challenging task because of the complex traffic sign scene and the small size of the target. In recent years, a number of convolutional neural network (CNN) based object detection methods have brought great progress to traffic sign detection. Considering the still high false detection rate, as well as the high time overhead and computational overhead, the effect is not satisfactory. Therefore, we employ lightweight network model YOLO v5 (You Only Look Once) as our work foundation. In this paper, we propose an improved YOLO v5 method by using balances feature pyramid structure and global context block to enhance the ability of feature fusion and feature extraction. To verify our proposed method, we have conducted a lot of comparative experiments on the challenging dataset Tsinghua-Tencent-100K (TT100K). The experimental results demonstrate that the mAP@.5 and mAP@.5:0.95 are improved by 1.9% and 2.1%, respectively.

Refined single-stage object detection deep-learning technique for chilli leaf disease detection

“…The proposed SSOD model has incorporated the Mosaic data augmentation technique to enhance the training images. 39 This method involves combining four separate images by applying random scaling, cutting, and layout to generate a single, mosaic image. The result is an enrichment of the background information in the training image, which greatly improves the ability to detect small targets.…”

“…The proposed SSOD model has incorporated the Mosaic data augmentation technique to enhance the training images 39 . This method involves combining four separate images by applying random scaling, cutting, and layout to generate a single, mosaic image.…”

Performance Analysis of Deep Learning YOLO models for South Asian Regional Vehicle Recognition

“…www.ijacsa.thesai.org Mean average precision is a well-known and commonly used object detection evaluation metric. Faster R-CNN [35], YOLO [36], and MobileNet [37] are all state-of-the-art models that use mAP to evaluate their models. We tried to test the performance of the three YOLOv5 modelssmall, medium, and largein our implementation.…”

Lightweight target detection algorithm based on improved YOLOv4

“…Therefore, the method proposed in this paper uses the Mish activation function to optimize the h-swish activation func-tion in the backbone feature extraction network. The swish x is calculated as shown in Equation (3).…”

“…As an important problem in computer vision, object detection is widely used in face detection [1,2], traffic sign detection [3,4], remote sensing image detection [5,6], big data [7,8], and other related fields. With the great progress of deep learning on image classification tasks, deep learning-based target detection algorithms have gradually become main stream.…”