In this paper, we introduce a video-based wildfire detection scheme based on a computationally efficient additive deep neural network, which we call AddNet. This AddNet is based on a multiplication-free vector operator, which performs only addition and sign manipulation operations. In this regard, we construct a dot product-like operation from the mf-operator and use it to define dense and convolutional feed-forwarding passes in AddNet. We train AddNet on images taken from forestry surveillance cameras. Our experiments show that AddNet can achieve a time-saving by 12.4% when compared to an equivalent regular convolutional neural network (CNN). Furthermore, the smoke recognition performance of AddNet is as good as regular CNNs and substantially better than binary-weight neural networks.

中文翻译：

---

用于森林火灾检测的加性神经网络

在本文中，我们介绍了一种基于视频的野火检测方案，该方案基于计算效率高的加性深度神经网络，我们称之为 AddNet。此 AddNet 基于无乘法向量运算符，该运算符仅执行加法和符号操作操作。在这方面，我们从 mf-operator 构建了一个类似点积的操作，并使用它来定义 AddNet 中的密集和卷积前馈通道。我们在从林业监控摄像机拍摄的图像上训练 AddNet。我们的实验表明，与等效的常规卷积神经网络 (CNN) 相比，AddNet 可以节省 12.4% 的时间。此外，AddNet 的烟雾识别性能与常规 CNN 一样好，并且明显优于二进制权重神经网络。