Current Issue : July - September Volume : 2020 Issue Number : 3 Articles : 5 Articles
Nowadays, computer vision as an interdisciplinary field is growing in different\nareas such as medical, electronics, etc. In the field, detection and particularly\nimage segmentation is an essential task in which is difficult to find the\nappropriate one based on the application. In this paper, a new algorithm is\nproposed to segment the lesion from background. The algorithm is based on\nlog edge detector with iterative median filtering. We have tested our algorithm\non 20 dermoscopic images and compare the lesion detection results\nwith those manually segmented by dermatologists. The experiments represent\nthe effectiveness of proposed algorithm....
Abnormal falls in public places have significant safety hazards and can easily lead to serious\nconsequences, such as trampling by people. Vision-driven fall event detection has the huge advantage\nof being non-invasive. However, in actual scenes, the fall behavior is rich in diversity, resulting in\nstrong instability in detection. Based on the study of the stability of human body dynamics, the\narticle proposes a new model of human posture representation of fall behavior, called the â??five-point\ninverted pendulum modelâ?, and uses an improved two-branch multi-stage convolutional neural\nnetwork (M-CNN) to extract and construct the inverted pendulum structure of human posture in\nreal-world complex scenes.................................
Gesture recognition is topical in computer science and aims at interpreting\nhuman gestures via mathematical algorithms. Among the numerous applications\nare physical rehabilitation and imitation games. In this work, we suggest\nperforming human gesture recognition within the context of a serious imitation\ngame, which would aim at improving social interactions with teenagers\nwith autism spectrum disorders. We use an artificial intelligence algorithm to\ndetect the skeleton of the participant, then model the human pose space and\ndescribe an imitation learning method using a Gaussian Mixture Model in the\nRiemannian manifold....
Die-stacking technology is expanding the space diversity of on-chip communications by\nleveraging through-silicon-via (TSV) integration and wafer bonding. The 3D network-on-chip\n(NoC), a combination of die-stacking technology and systematic on-chip communication\ninfrastructure, suffers from increased thermal density and unbalanced heat dissipation across multistacked\nlayers, significantly affecting chip performance and reliability......................
This study proposes a synthetic aperture radar (SAR) target-recognition method based on the fused features from the multiresolution\nrepresentations by 2D canonical correlation analysis (2DCCA). The multiresolution representations were demonstrated\nto be more discriminative than the solely original image. So, the joint classification of the multiresolution representations is\nbeneficial to the enhancement of SAR target recognition performance. 2DCCA is capable of exploiting the inner correlations of\nthe multiresolution representations while significantly reducing the redundancy. Therefore, the fused features can effectively\nconvey the discrimination capability of the multiresolution representations while relieving the storage and computational burdens\ncaused by the original high dimension. In the classification stage, the sparse representation-based classification (SRC) is employed\nto classify the fused features. SRC is an effective and robust classifier, which has been extensively validated in the previous works.\nThe moving and stationary target acquisition and recognition (MSTAR) data set is employed to evaluate the proposed method.\nAccording to the experimental results, the proposed method could achieve a high recognition rate of 97.63% for the 10 classes of\ntargets under the standard operating condition (SOC). Under the extended operating conditions (EOC) like configuration\nvariance, depression angle variance, and the robustness of the proposed method are also quantitively validated. In comparison\nwith some other SAR target recognition methods, the superiority of the proposed method can be effectively demonstrated....
Loading....