Energy-efficient secured data reduction technique using image difference function in wireless video sensor networks

Abstract

HAL is a multi-disciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L'archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d'enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Image Similarity Based Data Reduction Technique in Wireless Video Sensor Networks for Smart Agriculture

“…In this section, we discuss some previous works regarding data reduction in WSN and in WVSN in particular. A lot of techniques based on scheduling [6], [3], in-trusion location and clustering [13,4], compressive sensing [16], data similarity [10], [11], and frame rate adaptation [15,7] exist for data reduction and for energyefficiency. The authors in [13] used the kinematics functions to predict the next location of the intrusion in the area of interest in order to increase the frame rate adaptation of the targeted nodes.…”

“…This approach can come as a complementary solution to our method which focuses on the nodes detecting the intrusion and not on the projected positions of the intrusion. In [11] and [15], the authors worked on data reduction by adapting the frame rate of the nodes. This type of approaches is helpful and could be studied later on as a complementary work of the approach presented in this paper.…”

“…1. TDR algorithm leverages the consecutive frames similarity as in [11], using the norm simple euclidean distance similarity method to compare images. If two consecutive sensed frames are estimated as similar (no new information is represented in the second frame), the node does not send the second frame to the coordinator.…”

“…The node counts the number of consecutive similar frames, if it surpasses nb (the required number of consecutive similar frames), the state of the area of interest monitored by the node is considered as stable (situation=1). This technique uses the norm L2 relative error function tested in C++ in OpenCV and compared to other techniques in [11] where the time to execute this technique is way less important than other techniques (edge, colour, ...) for the same purpose of data reduction. To apply this function, each image is transformed into a matrix.…”

“…The second one, the Inter-Nodes Similarity algorithm (INS), leverages the spatio-temporal correlation between neighbour nodes to reduce the number of sent and captured frames. The TDR algorithm compares consecutive captured frames on the node based on a norm simple euclidean distance similarity method [11]. According to the degree of similarity with the last frame sent to the sink, the node decides whether to send this frame or not [12].…”