ECRU: An Encoder-Decoder Based Convolution Neural Network (CNN) for Road-Scene Understanding

Abstract

This research presents the idea of a novel fully-Convolutional Neural Network (CNN)-based model for probabilistic pixel-wise segmentation, titled Encoder-decoder-based CNN for Road-Scene Understanding (ECRU). Lately, scene understanding has become an evolving research area, and semantic segmentation is the most recent method for visual recognition. Among vision-based smart systems, the driving assistance system turns out to be a much preferred research topic. The proposed model is an encoder-decoder that performs pixel-wise class predictions. The encoder network is composed of a VGG-19 layer model, while the decoder network uses 16 upsampling and deconvolution units. The encoder of the network has a very flexible architecture that can be altered and trained for any size and resolution of images. The decoder network upsamples and maps the low-resolution encoder’s features. Consequently, there is a substantial reduction in the trainable parameters, as the network recycles the encoder’s pooling indices for pixel-wise classification and segmentation. The proposed model is intended to offer a simplified CNN model with less overhead and higher performance. The network is trained and tested on the famous road scenes dataset CamVid and offers outstanding outcomes in comparison to similar early approaches like FCN and VGG16 in terms of performance vs. trainable parameters.

Delimitación automática de ceniza volcánica en imágenes satelitales mediante Deep Learning

“…5) está establecido de acuerdo con el número de filtros (8, 16, 32 y 64) aplicados por capa; así el encoder se compone por 5 capas convolucionales, con filtros de 3x3 que crean mapas de características que capturan patrones en la imagen, en la que el volumen de salida está determinado según el número de filtros. A cada capa le sigue una unidad lineal rectificada (ReLU) que anula los valores negativos y deja pasar a los positivos a la siguiente capa tal como entran (Yasrab, 2018). La operación de MaxPooling de 2x2 realiza un muestreo descendente (downsampling), en donde, en cada contracción las dimensiones son reducidas a la mitad.…”

Convolutional-Neural-Network Assisted Segmentation and SVM Classification of Brain Tumor in Clinical MRI Slices

“…To achieve this task, the classifiers existing in the literature, such as SoftMax, SVM with various kernels (Linear, RBF, and Cubic) [3,12,33,[35][36][37], DA (Linear, and Quadratic) [12,33] and KNN (Fine, and Cubic) [12,33] are employed to accomplish the task. The earlier research also presents the similar medical image assessment tasks which implemented classifiers [38][39][40][41].…”

Autosegmentation of Prostate Zones and Cancer Regions from Biparametric Magnetic Resonance Images by Using Deep-Learning-Based Neural Networks

“…In the conventional SegNet, a rectified linear unit (ReLU) activation function is used in the convolution block; however, we replaced ReLU with an exponential linear unit (ELU) [19]. Many experiments have been performed with different activation functions, and ELU has been found to exhibit the best performance in this architecture [20].…”