In this study, we propose a deep learning based vehicle make and model recognition method for weakly illuminated near-infrared camera images (NIR). Unlike earlier approaches that consider color images obtained in well-lit environments, this study proposes an approach for images captured at night-time. In the proposed approach, vehicle localization is carried out using single shot multi-box detector (SSD)model. Next, we utilize a convolutional neural network (CNN)based vehicle model classifier on the detected vehicle region. Data sets of different shade, light, reflection and other lighting effects have been created to be used in the training and testing stages of the proposed methods. Vehicle make and model classification model was tested using 3327 real-world night-time NIR images collected on a roadway. In order to observe the performance of the proposed vehicle make-model recognition method on enhanced images, three popular low-light image enhancement methods are also applied to our test dataset. Proposed model achieved 86% accuracy rate in model recognition and 95% accuracy rate in make recognition tasks.

The increased use of digital video and image processing technology has paved the way for extending the traffic enforcement applications to a wider range of violations as well as making the enforcement process more efficient. Automated traffic enforcement has mainly been applied towards speed and red light violations detection. In recent years, there has been an extension to other violation detection tasks such as seat-belt usage, tailgating and toll payment violations. In the recent years, automated driver cell phone usage violation detection methods have aroused considerable interest since it results in higher mortality rates than the intoxicated driving. In this study, we propose a novel automated technique towards driver’s phone usage violation detection using deep learning algorithms. Using an existing license plate recognition camera system placed on an overhead gantry, installed on a highway, real world images are captured during day and night time. We performed experiments using more than 5900 real world images and achieved an overall accuracy of 90.8 % in the driver cell phone usage violation detection task.

Vehicle body damage detection from still images has received considerable interest in the computer vision community in recent years. Existing methods are typically developed towards the auto insurance industry to minimize the claim leakage problem. Earlier studies utilized images taken from short proximity (< 3 meters) to the vehicle or to the damaged region of vehicle. In this study, we investigate the vehicle frontal body damage detection using roadway surveillance camera images. The proposed method utilizes deep learning based object detection and image classification methods to determine damage status of a vehicle. The proposed method combines the symmetry property of vehicles’ frontal view and transfer learning concept in its inference process. Experimental results show that the proposed method achieves 91 % accuracy on a test dataset.

Manual seat belt enforcement by roadside police officers on highways is known to be inefficient, costly and ineffective. Authorities have been looking for automated enforcement solution towards seat belt violation detection. In this study, we examine different approaches towards the seat belt violation of front seat passengers using a near infrared (NIR) and color (RGB) surveillance camera system pointed at the vehicle's windshield. In these approaches an object detector; single shot multi box object detector (SSD) and image classifiers; convolutional neural network (CNN), Fisher vector (FV) are utilized to detect the usage of seat belt. Using a camera system placed on an overhead gantry installed on a highway, 13444 real world images are captured in a 20 hour time period. Our experiments conducted on this dataset show that SSD seat belt detector is superior for this problem with 91.9 % accuracy and 94.5 % precision rate.

Camera based traffic enforcement solutions has been ubiquitously employed in many roadways around the world. Toll violation detection, red light violation detection, speed violation detection are typically the most common usages of camera systems towards traffic enforcement. However, there are other enforcement areas that need attention such as front seat child occupancy violation, which is a major safety concern in many countries. In this study, we propose a novel automated technique towards child occupancy detection in front seat of vehicles using deep learning algorithms. Using a camera system placed on an overhead gantry, installed on a high way in Turkey, real world images are captured during day and night time. We performed experiments using 1000 real world images and achieved an overall accuracy of 92%.

Skin prick test is a common diagnosis method for allergic diseases. The test causes a controlled allergic response and provokes a wheal on the skin, whose size and shape change depending on the sensitivity of a patient against a specific allergen. The physical size of the wheal can be estimated from images acquired by digital cameras, which requires segmentation of the wheal from the captured SPT images. In this paper, we propose a new method for wheal detection from skin prick test images. The proposed method operates on localized images using a set of calibration marks and a perspective transformation. An initial segmentation of the localized SPT image is performed using normalized cuts to identify coherent regions in the image. The segmented region containing the wheal is selected by formulating an optimization problem to identify the segment with distinctive color, texture, and shape attributes. Our experiments conducted on images acquired from 50 different patients show the performance of the proposed wheal detection algorithm.

Due to the high volume of traffic on modern roadways, transportation agencies have proposed high occupancy vehicle (HOV) and high occupancy tolling (HOT) lanes to promote carpooling. Enforcement of the rules of these lanes is currently performed by roadside enforcement officers using visual observation. Officer-based enforcement is, however, known to be inefficient, costly, potentially dangerous, and ultimately ineffective. Violation rates up to 50%-80% have been reported, whereas manual enforcement rates of less than 10% are typical. Near-infrared (NIR) camera systems have been recently proposed to monitor HOV/HOT lanes and enforce the regulations. These camera systems bring an opportunity to automatically determine vehicle occupancy from captured HOV/HOT NIR images. Due to their ability to see through windshields of vehicles, these cameras also enable enforcement of other passenger compartment violations such as seatbelt violation and driver cell phone usage, in addition to determining vehicle occupancy. In this paper, we propose computer vision methods for detecting vehicle occupancy, seatbelt violation, and driver cell phone usage from NIR images captured from HOV/HOT lanes. Our methods consist of two stages. First, we localize the vehicle's front windshield and side window from captured HOV/HOT images using the deformable part model (DPM). Next, we define a region of interest in the localized images for each violation type and perform image classification using one of the local aggregation-based image features, i.e., bag-of-visual-words (BOW), vector of locally aggregated descriptors (VLAD), and Fisher vectors (FV), and compare their performances for each case. We also compare the performance of DPM-based detection with the image classification methods for vehicle occupancy and seatbelt violation detection. A data set over 4000 images including front/side view vehicle images with seatbelt and cell phone violations was collected on a public roadway and is used to perform the experiments.

Environmental pollution has recently aroused considerable public attention around the world. Air pollution is one form of pollution that is caused by the release of the chemical gases and particulates to the atmosphere. Remote monitoring of the chemical gas releasing facilities is urgently needed to regulate the amount of gas release levels. Hyperspectral imaging in long wave infrared (LWIR) spectrum is a popular way to monitor gas release in chemical facilities. In this study, we propose a novel approach towards semi-supervised detection/classification of gases in hyperspectral images. In our experiments, proposed approach is tested on 6 hyperspectral images and it achieves an average accuracy of 0.967.

Due to the high volume of traffic on modern roadways, transportation agencies have proposed High Occupancy Vehicle (HOV) lanes and High Occupancy Tolling (HOT) lanes to promote car pooling. However, enforcement of the rules of these lanes is currently performed by roadside enforcement officers using visual observation. Manual roadside enforcement is known to be inefficient, costly, potentially dangerous, and ultimately ineffective. Violation rates up to 50%–80% have been reported, while manual enforcement rates of less than 10% are typical. Therefore, there is a need for automated vehicle occupancy detection to support HOV/HOT lane enforcement. A key component of determining vehicle occupancy is to determine whether or not the vehicle's front passenger seat is occupied. In this paper, we examine two methods of determining vehicle front seat occupancy using a near infrared (NIR) camera system pointed at the vehicle's front windshield. The first method examines a state-of-the-art deformable part model (DPM) based face detection system that is robust to facial pose. The second method examines state-of-the-art local aggregation based image classification using bag-of-visual-words (BOW) and Fisher vectors (FV). A dataset of 3000 images was collected on a public roadway and is used to perform the comparison. From these experiments it is clear that the image classification approach is superior for this problem.

Prostate cancer localization using supervised classification techniques has aroused considerable interest in medical imaging community in recent years. However, it is crucial to have an accurate training data set for supervised classification techniques. Since different devices with, e.g., different protocols and/or field strengths cause different intensity profiles, each device/protocol must have an accompanying training data set, which is very costly to obtain. It is highly desirable to adapt the existing classifier(s) trained for one device/protocol to help classify data coming from another device/protocol. In this paper, we propose a novel method that has the ability to design classifiers obtained from one imaging protocol and/or MRI device to be used on a data set from another protocol and/or imaging device. As an example problem, we consider prostate cancer localization with multiparametric MRI. We show that simple normalization techniques such as z-score are not sufficient for cross-device automated cancer localization. On the other hand, the method we have originally developed based on relative intensity allows us to successfully use a classifier obtained from one device to be applied on a test patient imaged with another device. Proposed method also allows us to employ T2-weighted MR images directly instead of an additional step to normalize T2-weighted images usually performed in an ad hoc manner when T2 maps are not available. To demonstrate the effectiveness of the proposed method, we use a multiparametric MRI data set acquired from 18 biopsy-confirmed cancer patients with two separate scanners: 1) 1.5-T (Excite HD) GE and 2) 1.5-T (Achieva) Philips Healthcare scanners. A comprehensive visual, quantitative, and statistical analysis of the results show that methods we have developed allow us to: 1) perform cross-device automated classification and 2) use T2-weighted images without an ad hoc subject-specific normalization.

In this paper, we propose a novel and efficient semisupervised technique for automated prostate cancer localization using multiparametric magnetic resonance imaging (MRI). This method can be used in guiding biopsy, surgery, and therapy. We systematically present a new segmentation technique by developing a multiparametric graph based random walker (RW) algorithm with automated seed initialization to perform prostate cancer segmentation using multiparametric MRI. RW algorithm has proved to be accurate and fast in segmentation applications; however it requires a set of (user provided) seed points in order to perform segmentation. In this study, we first developed a novel RW method, which can be used with multiparametric MR images and then devised alternative methods that can determine seed points in an automated manner using discriminative classifiers such as support vector machines (SVM). Proposed RW method with automated seed initialization is able to produce improved segmentation results by assigning more weights to the images with more discriminative power.We applied the proposed method to a multiparametric dataset obtained from biopsy confirmed prostate cancer patients. Proposed method produces a sensitivity/ specificity rate of 0.76 and 0.86, respectively. Both visual, quantitative as well as statistical results are presented to show the significant performance improvements. Fisher sign test is used to demonstrate the statistical significance of our results by achieving p-values less than 0.05. This method outperforms available RW and SVM based methods by achieving a high specificity rate while not reducing sensitivity.

Since sensor longevity is crucial in wireless sensor networks, transmission saving is highly important. Ordering based energy efficient classification in sensor networks has shown significant savings in transmissions. The study presented in this paper improves upon the earlier work on ordered classification by extending the classification scheme to nonlinear kernel methods and introducing ordering based Neyman-Pearson (NP) classification. Given a specified level α ∈(0, 1), NP criterion ensures to keep a false alarm rate no greater than α while minimizing the miss rate. This study demonstrates transmissions can be saved, without degradation in error probability, using an ordering based NP-classification approach. The average number of transmissions saved is lower bounded by a quantity proportional to the number of sensors employed.

Prostate cancer is a leading cause of cancer death for men in the United States. Fortunately, the survival rate for early diagnosed patients is relatively high. Therefore, in vivo imaging plays an important role for the detection and treatment of the disease. Accurate prostate cancer localization with noninvasive imaging can be used to guide biopsy, radiotherapy, and surgery as well as to monitor disease progression. Magnetic resonance imaging (MRI) performed with an endorectal coil provides higher prostate cancer localization accuracy, when compared to transrectal ultrasound (TRUS). However, in general, a single type of MRI is not sufficient for reliable tumor localization. As an alternative, multispectral MRI, i.e., the use of multiple MRI-derived datasets, has emerged as a promising noninvasive imaging technique for the localization of prostate cancer; however almost all studies are with human readers. There is a significant inter and intraobserver variability for human readers, and it is substantially difficult for humans to analyze the large dataset of multispectral MRI. To solve these problems, this study presents an automated localization method using cost-sensitive support vector machines (SVMs) and shows that this method results in improved localization accuracy than classical SVM. Additionally, we develop a new segmentation method by combining conditional random fields (CRF) with a cost-sensitive framework and show that our method further improves cost-sensitive SVM results by incorporating spatial information. We test SVM, cost-sensitive SVM, and the proposed cost-sensitive CRF on multispectral MRI datasets acquired from 21 biopsy-confirmed cancer patients. Our results show that multispectral MRI helps to increase the accuracy of prostate cancer localization when compared to single MR images; and that using advanced methods such as cost-sensitive SVM as well as the proposed cost-sensitive CRF can boost the performance significantly when compared to SVM.

Classification employing sensors connected by wireless networks is of great interest. As the sensor nodes are usually powered by batteries, saving transmissions is important. We demonstrate transmissions can be saved, without degradation in error probability, using an ordering approach. The average number of transmissions saved (ANTS) is lower bounded by a quantity proportional to the number of sensors employed provided a well-behaved distance measure between the sensor distributions is sufficiently large. For such cases, the ANTS over the optimum unconstrained energy approach is shown to be larger than half the number of sensors employed.