Smart logistics using convolutional neural networks and sensor data fusion

Many transportation companies provides logistics services for transporting perishable material using its refrigerated trucks. They have a problem of large penalties being imposed because of delay in material delivery, material damage and material being stolen. Till date, there are no defined solutions for monitoring all these problems, but due to the advancements in communication technologies, cloud services, big data technologies, visualizing techniques and Internet of things [IOT] it is very well possible to determine if there is any theft attempt or any variation in the desired ambient conditions in the truck. The desired environmental conditions include, temperature, humidity, lighting and detection of human intervene. Also, the surveillance system is trained using Computational Neural Networks (CNN) by introducing a large number of sample images of theft attempts. The CNN algorithm classifies the input images based on the sample images.