China
Africa
Explore chapters and articles related to this topic
Singular Value Decomposition–Principal Component Analysis-Based Object Recognition Approach
Published in D. P. Acharjya, V. Santhi, Bio-Inspired Computing for Image and Video Processing, 2018
Chiranji Lal Chowdhary, D.P. Acharjya
A new feature extraction method for object recognition tasks based on singular value decomposion (SVA) and principal component analysis (PCA) has been proposed. Although PCA is intrinsically very close, SVD differs in important aspects that can affect the performance of the classifier used in the recognition system. The result is compared with other standard transforms, like PCA and ICA. In the proposed work, the appearance-based 3DOR techniques is studied and compared in equal working conditions regarding preprocessing and algorithm implementation, using a COIL-100 dataset, with its standard sets. The potential of SVD-PCA on a database of 7200 images of 100 different color objects is illustrated. Overall, it is observed that SVD-PCA performs significantly better than conventional and subsequent eigenvalue decompositions. Experimental results in appearance-based object recognition confirm that SVD-PCA offers better recognition rates over ICA and PCA. The excellent recognition rates achieved in all of the experiments performed indicates that the proposed method is well-suited for 3D object recognition in applications like surveillance, robot vision, biometrics, and security tasks, etc.
Mobile Robot Navigation Using an Object Recognition Software with RGBD Images and the YOLO Algorithm
Published in Applied Artificial Intelligence, 2019
Douglas Henke Dos Reis, Daniel Welfer, Marco Antonio De Souza Leite Cuadros, Daniel Fernando Tello Gamarra
There is another work where is created a modified version of YOLO algorithm aiming a better efficiency in 3D object recognition (Zhao, Jia, and Ni 2018). One of the modifications is that the bounding box becomes the cluster box. The main function of the cluster box is to encompass all the object inside its borders. The M-YOLO (as is called the modified YOLO) do not use depth information to build and recognize the 3D object, the recognition and the training is still based in 2D images, and the cluster box in 2D coordinates are mapped and transformed to 3D coordinates. The Zhao’s work shows another application for the YOLO algorithm that can be turned into a powerful 3D object recognition algorithm.