REAL-TIME OBJECT DETECTING USING
SPIKING CONVOLUTIONAL NEURAL NETWORK
In this research, a real-time object detection was planned to be processed using a Spiking Convolutional Neural Network (SCNN) with temporal coding into the AlexNet architecture with a novel adaptive chimp optimiser for real-time object detection. This novel Adaptive spiking AlexNet convolutional chimp optimiser is the spiked-based real-time object detection model that provides near-lossless information transmission in a shorter period of time for deep SCNN. The primary layer of this SCNN consists of dissimilar Gaussian filters to predict the contrast of the video frames from the real-time video datasets. Hereafter, it encodes the contrast strength over the latencies. The design of spikes in the convolutional model has support to find the complex features of the video frames. Also, it can take the output of previous layers as input data for the following layers. When the object has been detected simultaneously Convolutional neurons emit a spike, which is dependent on input weight. The translation invariance has helped to minimize the data size. Pooling Layer neurons broadcast the initial spike that has received from nearer previous layer neurons but it has selectively the same features. Finally, the object feature present in the video has been detected.
§ Initially, real-time videos are collected from the net sources and trained to the system
§ Hereafter a spiking convolutional neural network has been designed in the python framework
§ Then the videos are separated into the number of frames
§ The Gaussian filter in the neural network has eliminated the training error, after the pre-processing, the cleaned data was imported to the pooling layer
§ The features of the trained video was detected using the spike properties in the pooling layer
§ Finally, the key metrics are calculated to report the efficiency measure of the proposed model.
12 freelancer chào giá trung bình₹53125 cho công việc này
Hi, I have 6 years of programming experience in Python, and 4 in machine learning, both CV and NLP. Please, tell, which dataset will you use? What is your budget and deadlines?