We present a model for real-time pedestrian detection based on a deep learning framework. With respect to the base network for feature extraction, we have improved the network based on Mobilenet which is a simple and fast convolutional neural network. We only use the front part of its network and then build several new multi-scale convolutional layers to calculate multi-scale feature maps. With respect to the detection network behind the feature extraction, we use a simplified SSD(single shot multibox detector) model to detect pedestrians with fewer feature maps. In addition, we design detection boxes with specific sizes according to pedestrian’s shape characteristics. To avoid overfitting, we apply data augmentation and dropout techniques to training. Experimental results on PASCAL VOC and KITTI confirm that the speed of our detection model has been increased by 22.2% while precision remains almost unchanged. Our approach makes a trade-off between speed and precision, and has an obvious speed advantage over other detection approaches.
We present a new approach to face detection with skin color mixture models and asymmetric AdaBoost. First, non-skin
color pixels of the input image are rapidly removed based on skin color mixture models in RGB and YCbCr
chrominance spaces, from which we extract candidate face regions. Then, face detection with fast asymmetric AdaBoost
is carried out in candidate face regions where ratios of pixels of skin color to non-skin color are beyond certain
thresholds. To further reduce the computational cost, the integral image technique is employed to calculate ratios of
pixels of skin color to non-skin color in candidate face regions. Finally, false alarms are gradually merged and removed
by relative geometric relation and the rate of skin color pixels on the intersection line of candidate face regions.
Experimental results show that our proposed method reduces significantly false alarms and the processing time while
achieves detection rates of more than 99%.
Automatic object recognition is an active area of interest for military and commercial applications. In this paper, a
system addressing autonomous recognition of ship types in infrared images is proposed. Firstly, an approach of
segmentation based on detection of salient features of the target with subsequent shadow removing is proposed, as is the
base of the subsequent object recognition. Considering the differences between the shapes of various ships mainly lie in
their superstructures, we then use superstructure moment functions invariant to translation, rotation and scale differences
in input patterns and develop a robust algorithm of obtaining ship superstructure. Subsequently a back-propagation
neural network is used as a classifier in the recognition stage and projection images of simulated three-dimensional ship
models are used as the training sets. Our recognition model was implemented and experimentally validated using both
simulated three-dimensional ship model images and real images derived from video of an AN/AAS-44V Forward
Looking Infrared(FLIR) sensor.
The infrared ship segmentation in digital images is a fundamental step in the process of ship recognition. This paper
presents an adaptive recursive algorithm for infrared ship image segmentation based on the gray-level histogram analysis
of the image. The proposed algorithm consists of four phases. First, the gray-level histogram of the image is generated
and de-noised by using wavelets transform. Second, a threshold level which best extracts the ship from the water region
is selected according to the histogram profile analysis. Third, the rationality of the selected threshold is analyzed based
on the prior information about infrared ship images. If the selected threshold is not reasonable, we can still use it as the
recursive initial threshold and the infrared ship image will be further segmented with a local recursive method based on
the method proposed by OTSU until it reaches the prescriptive termination criteria. Finally, we eliminate the spurious
pixels by extracting the greatest connected region and filling the holes. The segmentation algorithm works successfully
for classification of infrared ships, and some experimental results are also presented.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.