Reduced computational cost prototype for street theft detection based on depth decrement in Convolutional Neural Network. Application to Command and Control Information Systems (C2IS) in the National Police of Colombia.

Abstract

This paper shows the implementation of a prototype of street theft detector using the deep learning technique R-CNN (Region-Based Convolutional Network), applied in the Command and Control Information System (C2IS) of National Police of Colombia, the prototype is implemented using three models of CNN (Convolutional Neural Network), AlexNet, VGG16 and VGG19 comparing their computational cost measuring the image processing time, according to the complexity (depth) of each model. Finally, we conclude which model has the lowest computational cost and is more useful for the case of the National Police of Colombia.

A Novel Low Processing Time System for Criminal Activities Detection Applied to Command and Control Citizen Security Centers

“…Regarding the system accuracy, we checked that it is free of overtraining as the tests done on the system were performed with images not used in the training process and their results were confirmed in the confusion matrix and the system accuracy it is within the range expected for a Faster R-CNN; however, the system is designed to be used in public safety applications, so it always requires human monitoring because the detections depend on the lighting conditions and the distance of the cameras to the object, in addition to the success rate of the Faster R-CNN; additionally, in previous studies [22], authors evaluated other CNN models of a greater depth by choosing AlexNet for its performance and simplicity.…”

“…To choose the Deep Learning Models, we studied factors such as the processing time of each video frame, accuracy and model robustness. Therefore, several detection techniques were studied, such as R-CNN (Region-Based Convolutional Network) [19], YOLO (You Only Look Once) [20], Fast R-CNN (Fast Region-Based Convolutional Network) [21,22] and Faster R-CNN (Faster Region-Based Convolutional Network) [23,24] (Table 1). After analyzing the advantages and disadvantages of each technique, Faster R-CNN was chosen to implement the system for criminal events detection in the system for the C2IS of the National Colombian Police due to the fact that it has an average timeout that was 250 times faster than R-CNN and 25 times faster than Fast R-CNN [22,25,26].…”

“…Therefore, several detection techniques were studied, such as R-CNN (Region-Based Convolutional Network) [19], YOLO (You Only Look Once) [20], Fast R-CNN (Fast Region-Based Convolutional Network) [21,22] and Faster R-CNN (Faster Region-Based Convolutional Network) [23,24] (Table 1). After analyzing the advantages and disadvantages of each technique, Faster R-CNN was chosen to implement the system for criminal events detection in the system for the C2IS of the National Colombian Police due to the fact that it has an average timeout that was 250 times faster than R-CNN and 25 times faster than Fast R-CNN [22,25,26]. Furthermore, in recent work, models based on two stages like Faster R-CNN have had better accuracy and stability than models based on regression like YOLO [27,28] and SSD, which is of great importance because in this work, a novel application focused in action detection was given to an object detector model.…”

“…It has less layers than models like VGG16 [56], VGG19 [56], GoogleNet [57] or ResNet [58]. Hence, AlexNet has a lower computational cost and requires less processing time per video frame [22] (further implementation details are provided in Section 5). In this case, the pre-trained CNN model AlexNet [55] was used as the core of the RPN.…”

A Novel Method of Spatiotemporal Dynamic Geo-Visualization of Criminal Data, Applied to Command and Control Centers for Public Safety

“…Basic architecture of any C2S contains: Command and Control Communication Systems (or Subsystems) (C2CS), Command and Control Information Systems (or Subsystems) (C2IS), sensor systems and/or information sources; and command posts [16]. In the Command and Control Centers for Public Safety, the C2IS of the PONAL (which is where the tool that constitutes the method presented here, would fit); gathers the information from several sources of information, such as: institutional databases, sensor systems and video surveillance systems [46,47], among others, Figure 1; and Geographic Information Systems (GIS) to show relevant information from the aforementioned sources. They also have internal tools for processing data in order to extract more information.…”