Industrial Technology Research Institute

Tse-Min Chen

Peter Chondro

Pei-Jung Liang

Wei-Hao Lai

Wen-Shan Wang

Shanq-Jang Ruan

Chao-Yi Cho

The disclosure is directed to an object detection method using a CNN model and an object detection apparatus thereof. In an aspect, the object detection method includes generating a sensor data; processing the sensor data by using a first object detection algorithm to generate a first object detection result; processing the first object detection result by using a plurality of stages of sparse update mapping algorithm to generate a plurality of stages of updated first object detection result; processing a first stage of the stages of updated first object detection result by using a plurality of stages of spatial pooling algorithm between each of stages of sparse update mapping algorithm; executing a plurality of stages of deep convolution layer algorithm to extract a plurality of feature results; and performing a detection prediction based on a last-stage feature result.

Object detection method using CNN model and object detection apparatus using the same

A method for detecting dead pixels obtains a downsampling set formed by a plurality of sample pixels by downsampling a to-be-tested image, and computes a moving average for each sample pixel to determine whether the sample pixel is a dead pixel candidate. If so, all pixels in a neighborhood of the dead pixel candidate are determined as to-be-detected pixels. For each to-be-detected pixel, a compensated value and a weighted average under a mask are estimated. A hypothetic dead pixel value is obtained by comparing an original pixel value of each to-be-detected pixel and the moving average of the dead pixel candidate. Whether the to-be-detected pixel is a dead pixel is determined by using the original pixel value, the compensated value, the weighted average and the hypothetic dead pixel value.

Method for detecting dead pixels and computer program product thereof

A dynamic decoding lookup table generation method is provided. The generation method includes: receiving a variable length coding (VLC) table; and dynamically establishing an adaptive decoding lookup table from the VLC table according to a target decoding rate and a free system memory resource.

Growing community of inventors

Hsinchu, Taiwan

Tse-Min Chen