Showing papers on "Digital camera published in 2017"

DSLR-Quality Photos on Mobile Devices with Deep Convolutional Networks

Abstract: Despite a rapid rise in the quality of built-in smartphone cameras, their physical limitations – small sensor size, compact lenses and the lack of specific hardware, – impede them to achieve the quality results of DSLR cameras. In this work we present an end-to-end deep learning approach that bridges this gap by translating ordinary photos into DSLR-quality images. We propose learning the translation function using a residual convolutional neural network that improves both color rendition and image sharpness. Since the standard mean squared loss is not well suited for measuring perceptual image quality, we introduce a composite perceptual error function that combines content, color and texture losses. The first two losses are defined analytically, while the texture loss is learned in an adversarial fashion. We also present DPED, a large-scale dataset that consists of real photos captured from three different phones and one high-end reflex camera. Our quantitative and qualitative assessments reveal that the enhanced image quality is comparable to that of DSLR-taken photos, while the methodology is generalized to any type of digital camera.

DSLR-Quality Photos on Mobile Devices with Deep Convolutional Networks

Abstract: Despite a rapid rise in the quality of built-in smartphone cameras, their physical limitations - small sensor size, compact lenses and the lack of specific hardware, - impede them to achieve the quality results of DSLR cameras. In this work we present an end-to-end deep learning approach that bridges this gap by translating ordinary photos into DSLR-quality images. We propose learning the translation function using a residual convolutional neural network that improves both color rendition and image sharpness. Since the standard mean squared loss is not well suited for measuring perceptual image quality, we introduce a composite perceptual error function that combines content, color and texture losses. The first two losses are defined analytically, while the texture loss is learned in an adversarial fashion. We also present DPED, a large-scale dataset that consists of real photos captured from three different phones and one high-end reflex camera. Our quantitative and qualitative assessments reveal that the enhanced image quality is comparable to that of DSLR-taken photos, while the methodology is generalized to any type of digital camera.

Obstacle Detection and Avoidance System Based on Monocular Camera and Size Expansion Algorithm for UAVs

Abstract: One of the most challenging problems in the domain of autonomous aerial vehicles is the designing of a robust real-time obstacle detection and avoidance system. This problem is complex, especially for the micro and small aerial vehicles, that is due to the Size, Weight and Power (SWaP) constraints. Therefore, using lightweight sensors (i.e., Digital camera) can be the best choice comparing with other sensors; such as laser or radar.For real-time applications, different works are based on stereo cameras in order to obtain a 3D model of the obstacles, or to estimate their depth. Instead, in this paper, a method that mimics the human behavior of detecting the collision state of the approaching obstacles using monocular camera is proposed. The key of the proposed algorithm is to analyze the size changes of the detected feature points, combined with the expansion ratios of the convex hull constructed around the detected feature points from consecutive frames. During the Aerial Vehicle (UAV) motion, the detection algorithm estimates the changes in the size of the area of the approaching obstacles. First, the method detects the feature points of the obstacles, then extracts the obstacles that have the probability of getting close toward the UAV. Secondly, by comparing the area ratio of the obstacle and the position of the UAV, the method decides if the detected obstacle may cause a collision. Finally, by estimating the obstacle 2D position in the image and combining with the tracked waypoints, the UAV performs the avoidance maneuver. The proposed algorithm was evaluated by performing real indoor and outdoor flights, and the obtained results show the accuracy of the proposed algorithm compared with other related works.

Color image processing for digital cameras

Abstract: Color digital cameras are used by a growing number of consumers and professional photographers. These cameras use one or more charge-coupled-device or CMOS image sensors to capture color records of the scene, and they digitally process the color records to produce color image files. The professional cameras store unrendered image data, which later will be processed using proprietary software running on a separate host computer to complete the camera image processing. The capabilities and performance of a digital camera depend on both the camera’s hardware architecture and its image processing algorithms, often provided by firmware. Depending on the camera design, the camera image processing may be performed almost entirely in the camera application-specific integrated circuit or, alternatively, by a microprocessor incorporating a digital signal processor designed to provide rapid image processing. The tone scale/color reproduction process also must account for the fact that the dynamic range of a rendered image usually is substantially less than that of an original scene.

An innovative silicon photomultiplier digitizing camera for gamma-ray astronomy

Abstract: The single-mirror small-size telescope (SST-1M) is one of the three proposed designs for the small-size telescopes (SSTs) of the Cherenkov Telescope Array (CTA) project. The SST-1M will be equipped with a 4 m-diameter segmented reflector dish and an innovative fully digital camera based on silicon photo-multipliers. Since the SST sub-array will consist of up to 70 telescopes, the challenge is not only to build telescopes with excellent performance, but also to design them so that their components can be commissioned, assembled and tested by industry. In this paper we review the basic steps that led to the design concepts for the SST-1M camera and the ongoing realization of the first prototype, with focus on the innovative solutions adopted for the photodetector plane and the readout and trigger parts of the camera. In addition, we report on results of laboratory measurements on real scale elements that validate the camera design and show that it is capable of matching the CTA requirements of operating up to high moonlight background conditions.

Development of photo forensics algorithm by detecting photoshop manipulation using error level analysis

Abstract: Nowadays, image manipulation is common due to the availability of image processing software, such as Adobe Photoshop or GIMP. The original image captured by digital camera or smartphone normally is saved in the JPEG format due to its popularity. JPEG algorithm works on image grids, compressed independently, having size of 8x8 pixels. For unmodified image, all 8x8 grids should have a similar error level. For resaving operation, each block should degrade at approximately the same rate due to the introduction of similar amount of errors across the entire image. For modified image, the altered blocks should have higher error potential compred to the remaining part of the image. The objective of this paper is to develop a photo forensics algorithm which can detect any photo manipulation. The error level analysis (ELA) was further enhanced using vertical and horizontal histograms of ELA image to pinpoint the exact location of modification. Results showed that our proposed algorithm could identify successfully the modified image as well as showing the exact location of modifications.

A Mobile Device-Based Imaging Spectrometer for Environmental Monitoring by Attaching a Lightweight Small Module to a Commercial Digital Camera

Abstract: Spatially-explicit data are essential for remote sensing of ecological phenomena. Lately, recent innovations in mobile device platforms have led to an upsurge in on-site rapid detection. For instance, CMOS chips in smart phones and digital cameras serve as excellent sensors for scientific research. In this paper, a mobile device-based imaging spectrometer module (weighing about 99 g) is developed and equipped on a Single Lens Reflex camera. Utilizing this lightweight module, as well as commonly used photographic equipment, we demonstrate its utility through a series of on-site multispectral imaging, including ocean (or lake) water-color sensing and plant reflectance measurement. Based on the experiments we obtain 3D spectral image cubes, which can be further analyzed for environmental monitoring. Moreover, our system can be applied to many kinds of cameras, e.g., aerial camera and underwater camera. Therefore, any camera can be upgraded to an imaging spectrometer with the help of our miniaturized module. We believe it has the potential to become a versatile tool for on-site investigation into many applications.

WiseEye: Next Generation Expandable and Programmable Camera Trap Platform for Wildlife Research

Abstract: The widespread availability of relatively cheap, reliable and easy to use digital camera traps has led to their extensive use for wildlife research, monitoring and public outreach. Users of these units are, however, often frustrated by the limited options for controlling camera functions, the generation of large numbers of images, and the lack of flexibility to suit different research environments and questions. We describe the development of a user-customisable open source camera trap platform named 'WiseEye', designed to provide flexible camera trap technology for wildlife researchers. The novel platform is based on a Raspberry Pi single-board computer and compatible peripherals that allow the user to control its functions and performance. We introduce the concept of confirmatory sensing, in which the Passive Infrared triggering is confirmed through other modalities (i.e. radar, pixel change) to reduce the occurrence of false positives images. This concept, together with user-definable metadata, aided identification of spurious images and greatly reduced post-collection processing time. When tested against a commercial camera trap, WiseEye was found to reduce the incidence of false positive images and false negatives across a range of test conditions. WiseEye represents a step-change in camera trap functionality, greatly increasing the value of this technology for wildlife research and conservation management.

Image style migration method based on deep convolutional neural network

Abstract: The invention discloses an image text description method based on a visual attention model. The main content comprises the followings: image inputting, loss function training, stylizing, image enhancing and image thinning; and the processes are as follows: an input image is firstly adjusted as a content image (256*256) with a dual-linear down-sampling layer, and then stylized through a style subnet; and then a stylized result as the first output image is up-sampled as an image in the size of 512*512, and then the up-sampled image is enhanced through an enhancement subnet to obtain the second output image; the second output image is adjusted as the image in the size of 1024*1024, and finally, a thinning subnet deletes locally pixelated artifact and further thins the result to obtain a high-resolution result. By use of the image style migration method disclosed by the invention, the brushwork of the artwork can be simulated more closely; multiple models are combined into a network so as to process the image with bigger and bigger size shot by a modern digital camera; and the method can be used for training the combined model to realize the migration of multiple artistic styles.

Smartphones as a Light Measurement Tool: Case of Study

Abstract: In recent years, smartphones have become the main computing tool for most of the population, making them an ideal tool in many areas. Most of these smartphones are equipped with cutting-edge hardware on their digital cameras, sensors and processors. For this reason, this paper discusses the possibility of using smartphones as an accessible and accurate tool, focusing on the measurement of light, which is an element that has a high impact on human behavior, which promotes conformance and safety, or alters human physiology when it is inappropriate. To carry out this study, three different ways to measure light through smartphones have been checked: the ambient light sensor, the digital camera and an external Bluetooth luxmeter connected with the smartphone. As a result, the accuracy of these methods has been compared to check if they can be used as accurate measurement tools.

Bridge Crack Detection Using Multi-Rotary Uav and Object-Base Image Analysis

Abstract: . Bridge is an important infrastructure for human life. Thus, the bridge safety monitoring and maintaining is an important issue to the government. Conventionally, bridge inspection were conducted by human in-situ visual examination. This procedure sometimes require under bridge inspection vehicle or climbing under the bridge personally. Thus, its cost and risk is high as well as labor intensive and time consuming. Particularly, its documentation procedure is subjective without 3D spatial information. In order cope with these challenges, this paper propose the use of a multi-rotary UAV that equipped with a SONY A7r2 high resolution digital camera, 50 mm fixed focus length lens, 135 degrees up-down rotating gimbal. The target bridge contains three spans with a total of 60 meters long, 20 meters width and 8 meters height above the water level. In the end, we took about 10,000 images, but some of them were acquired by hand held method taken on the ground using a pole with 2–8 meters long. Those images were processed by Agisoft PhotoscanPro to obtain exterior and interior orientation parameters. A local coordinate system was defined by using 12 ground control points measured by a total station. After triangulation and camera self-calibration, the RMS of control points is less than 3 cm. A 3D CAD model that describe the bridge surface geometry was manually measured by PhotoscanPro. They were composed of planar polygons and will be used for searching related UAV images. Additionally, a photorealistic 3D model can be produced for 3D visualization. In order to detect cracks on the bridge surface, we utilize object-based image analysis (OBIA) technique to segment the image into objects. Later, we derive several object features, such as density, area/bounding box ratio, length/width ratio, length, etc. Then, we can setup a classification rule set to distinguish cracks. Further, we apply semi-global-matching (SGM) to obtain 3D crack information and based on image scale we can calculate the width of a crack object. For spalling volume calculation, we also apply SGM to obtain dense surface geometry. Assuming the background is a planar surface, we can fit a planar function and convert the surface geometry into a DSM. Thus, for spalling area its height will be lower than the plane and its value will be negative. We can thus apply several image processing technique to segment the spalling area and calculate the spalling volume as well. For bridge inspection and UAV image management within a laboratory, we develop a graphic user interface. The major functions include crack auto-detection using OBIA, crack editing, i.e. delete and add cracks, crack attributing, 3D crack visualization, spalling area/volume calculation, bridge defects documentation, etc.

Digital micromirror device camera with per-pixel coded exposure for high dynamic range imaging

Abstract: In this paper, we overcome the limited dynamic range of the conventional digital camera, and propose a method of realizing high dynamic range imaging (HDRI) from a novel programmable imaging system called a digital micromirror device (DMD) camera. The unique feature of the proposed new method is that the spatial and temporal information of incident light in our DMD camera can be flexibly modulated, and it enables the camera pixels always to have reasonable exposure intensity by DMD pixel-level modulation. More importantly, it allows different light intensity control algorithms used in our programmable imaging system to achieve HDRI. We implement the optical system prototype, analyze the theory of per-pixel coded exposure for HDRI, and put forward an adaptive light intensity control algorithm to effectively modulate the different light intensity to recover high dynamic range images. Via experiments, we demonstrate the effectiveness of our method and implement the HDRI on different objects.

Small camera as a handheld colorimetric tool in the analytical chemistry

Abstract: Recently, there is an effort to introduce new types of analytical procedures and handheld assays to provide simple and reliable equipment for the field and household analyses. Development of diagnostic tools for self-diagnosis is another challenge in analytical chemistry. Digital cameras are widely available and cheap, hence they could be the sensor platform for construction of analytical and diagnostic methods. In general, good availability of cameras integrated into smartphones can be easily converted into an analytical tool. This review relates to the use of digital camera in analytical chemistry and there are introduced the facts how digital data can be processed and what the limits of digital photography are. Recent papers in this issue and discussion of development in camera based assays is also provided here.

A Widely Applicable Real-Time Mono/Binocular Eye Tracking System Using a High Frame-Rate Digital Camera

Abstract: We have developed a new mono/binocular eye tracking system by using an IEEE1394b or USB-3.0 digital camera that provides high sensitivity, high resolution, high frame-rate and no rolling shutter distortion. Our goal is to provide a system that is friendly to researchers who conduct experiments. The system is non-invasive and inexpensive and can be used for mice, marmosets, monkeys, and humans. It has adopted infrared light to illuminate an eye (eyes). The reflected image of the infrared light on the cornea and the black image of the pupil are captured by the camera. The center of the pupil and the center of the corneal reflection are calculated and tracked over time. The movement of the head is compensated by the reflection. Since the high resolution camera has a 2048 horizontal pixels resolution, we can capture the images of both eyes simultaneously and calculate the parameters of the two eyes at each frame. The gaze position data can be read out on-line via computer network and/or DAC (digital analog converter). The adoption of the Windows 10 as the operation system makes this eye tracking system user-friendly. Because of the high frame-rate of the digital camera, the sampling rate of the system can be as high as 700 Hz and the latency less than 4 ms.

Semi professional digital camera calibration techniques for Vis/NIR spectral data acquisition from an unmanned aerial vehicle

Abstract: Unmanned aerial vehicles UAVs equipped with multispectral digital cameras are very effective in spectral information obtainment. In order to check the radiometric response of the multispectral digital camera and provide better target’s spectral dynamics investigation, the used camera must be carefully calibrated in laboratory and field conditions. This article aimed to achieve the radiometric calibration of a multispectral Vis/NIR camera attached to an UAV, using as reference a hyperspectral sensor. First, a laboratory calibration, under controlled conditions, was performed. Then, a field calibration with no controlled conditions was accomplished. Finally, a cross calibration between laboratory and field data was developed. Multispectral Fujifilm S200-EXR digital camera, sensible to infrared spectrum radiation and 8 bits of radiometric resolution 256 digital numbers DNs and Fieldspec 3 Jr spectroradiometer ASD Inc., with spectral resolution of 3 nm between 350 and 1400 nm and 30 nm between 1400 and 2500 nm were used for spectral data acquisition. Six tarpaulins were used as reference targets. In laboratory, reference values were collected by Fieldspec 3 Jr and photographs of the reference targets and Spectralon panel were taken with no optical filter and using seven optical filters, which block the visible Vis and near-infrared NIR radiation in different intensities. In field, reflectance values were collected by the hyperespectral sensor on ground level and Vis/NIR images were taken using two identical Fujifilm S200-EXR cameras coupled to a Tarot Iron Man 1000 octocopter UAV at 200 and 600 m flight altitude on 22 January 2016 in clear weather conditions. All models and calibrations equations obtained by the correlation of DNs and reflectance values were significant to the Student’s t-test p ≤ 0.05. As in laboratory as well as in field, for red, green, blue, and NIR bands were obtained great correlations r > 0.90 except for red band in laboratory r = 0.88. The cross calibration laboratory vs field obtained models for red, green, blue, and NIR bands presented high Pearson coefficients r > 0.90. Under these circumstances, the calibrations models for red, green, blue, and NIR bands, point out the potential of cross calibration such way that the reference target’s reflectance values can be acquired in laboratory and, from digital images obtained by cameras attached to UAVs, DNs in field conditions can be collected, making able the easy obtainment of trusty spectral information.

IoT Based Home Security through Digital Image Processing Algorithms

Abstract: This paper gives an outline for automatic system to control and secure the home, based on digital image processing with the help of Internet of Things (IoT). The system consists of a sensor, digital camera, database in the fog and the mobile phone. Sensors are placed in the frame of the door which alerts camera, to capture an image who intends to enter the house, then sends the image to the database or dataset that is stored in the fog. Image analysis is performed to detect and recognize and match the image with the stored dataset of the authenticated people or pets. If the image captured does not match with the dataset then an alert message is send to the owner of the house. The image processing algorithms are considered for the processing spatial and time complexity of the image captured to cross check with the dataset stored in the fog.

Adaptive fringe projection technique for high-dynamic range three-dimensional shape measurement using binary search

Abstract: Fringe projection profilometry is a popular optical method for three-dimensional (3-D) shape measurement because of its high accuracy, fast measurement speed, and full-field inspection nature. However, due to the limited dynamic range of the digital camera, saturated pixels in the captured images will lead to serious phase errors and measurement errors when the measured object has a drastic texture variation. To deal with such a problem, an adaptive digital fringe projection technique for high dynamic 3-D shape measurement is proposed. In this method, phase-shifting fringes are adaptively generated with the aid of a coordinates mapping process and binary-search technique to eliminate saturation. Compared with previous adaptive fringe projection techniques, the camera response function and homographic mapping between the camera and projector are not needed, making the whole measurement easier to carry out and less laborious. Experiments validate the effectiveness and superiority of the proposed method for high-dynamic range 3-D shape measurement.

The effect of jpeg compression in close range photogrammetry

Abstract: Digital photogrammetry, using digital camera images, is an important low-cost engineering method to produce precise three-dimensional model of either an object or the part of the earth depending on the image quality. Photogrammetry which is cheaper and more practical than the new technologies such as LIDAR, has increased point cloud generation capacity during the past decade with contributions of computer vision. Images of new camera technologies needs huge storage space due to larger image file sizes. Moreover, this enormousness increases image process time during extraction, orientation and dense matching. The Joint Photographic Experts Group (JPEG) is one of the most commonly used methods as lossy compression standard for the storage purposes of the oversized image file. Particularly, image compression at different rates causes image deteriorations during the processing period. Therefore, the compression rates affect accuracy of photogrammetric measurements. In this study, the close range images compressed at the different levels were investigated to define the compression effect on photogrammetric results, such as orientation parameters and 3D point cloud. The outcomes of this study show that lower compression ratios are acceptable in photogrammetric process when moderate accuracy is sufficient.

Identification of mobile phones using the built-in magnetometers stimulated by motion patterns

Abstract: We investigate the identification of mobile phones through their built-in magnetometers. These electronic components have started to be widely deployed in mass market phones in recent years, and they can be exploited to uniquely identify mobile phones due their physical differences, which appear in the digital output generated by them. This is similar to approaches reported in the literature for other components of the mobile phone, including the digital camera, the microphones or their RF transmission components. In this paper, the identification is performed through an inexpensive device made up of a platform that rotates the mobile phone under test and a fixed magnet positioned on the edge of the rotating platform. When the mobile phone passes in front of the fixed magnet, the built-in magnetometer is stimulated, and its digital output is recorded and analyzed. For each mobile phone, the experiment is repeated over six different days to ensure consistency in the results. A total of 10 phones of different brands and models or of the same model were used in our experiment. The digital output from the magnetometers is synchronized and correlated, and statistical features are extracted to generate a fingerprint of the built-in magnetometer and, consequently, of the mobile phone. A SVM machine learning algorithm is used to classify the mobile phones on the basis of the extracted statistical features. Our results show that inter-model classification (i.e., different models and brands classification) is possible with great accuracy, but intra-model (i.e., phones with different serial numbers and same model) classification is more challenging, the resulting accuracy being just slightly above random choice.

On the applicability of inverse perspective mapping for the forward distance estimation based on the HSV colormap

Abstract: Advanced Driver Assistance System (ADAS) functionalities present in modern vehicles provide traffic improvement, comfort and safety for drivers, pedestrians, and the environment through the application of electronic, control and software. A digital camera is one of the technologies used in ADAS functions that can be installed in front of the vehicle. The camera acquires video and image in real time to measure the distance between the car and a point or object ahead. However, the problem of perspective effects can be seen in this domain, once the image acquired don't represents the real image due to distortions and errors in the processing stage. Some systems based on video processing require the ego vehicle to the target vehicle distance, as well as traffic signs and venerable obstacles such as pedestrians' distances. The method of Inverse Perspective Mapping (IPM) supports this demand by removing image distortions caused by perspective effect, generating new image coordinates in real-time. This paper presents the IPM and the implementation of a function to identify the distance from an object removing the perspective effect in the HSV colormap. The IPM method has been implemented in a real platform and tested in a vehicle under the road for the validation of its applicability.

Identification and Counting of Soybean Aphids from Digital Images Using Shape Classification

Abstract: . Soybean aphids are serious pests, causing negative yield impacts in the crop. Assessing their population is essential for making appropriate pesticide application decisions. Manual identification and counting, which is commonly performed to determine the economic threshold level, is time-consuming, laborious, and causes visual fatigue. In this study, an automatic image processing method was developed to identify and count aphids as well as exoskeletons and leaf spots on soybean leaves based on shape analysis. Aphid-infested soybean trifoliates were obtained at three infestation rates (low, medium, and high). Images of the front sides of the leaves were captured in the laboratory with three cameras (digital single-lens reflex or DSLR, consumer-grade digital, and smartphone) under two lighting conditions (direct and indirect). The shape parameters considered were area, perimeter, convex area, eccentricity, aspect ratio, solidity, hollowness, and roundness. Among the shape parameters tested, hollowness was the best in identifying aphids and was therefore used for developing the object classification algorithm. Of the three cameras tested, images from the consumer-grade digital camera produced the best identification accuracy (>82.4%), followed by the DSLR camera (>81.2%) and smartphone camera (>37.9%). Statistical analysis revealed that the accuracies did not differ significantly under different lighting conditions (p = 0.43), but the accuracies differed for the smartphone camera compared to the DSLR and consumer-grade digital cameras (p = 8.87 x 10-10). The results of automatic and manual counting were very well correlated (r = 0.92). The automatic image processing method achieved more rapid counting (

Camera calibration accuracy at different uav flying heights

Abstract: . Unmanned Aerial Vehicles (UAVs) can be used to acquire highly accurate data in deformation survey, whereby low-cost digital cameras are commonly used in the UAV mapping. Thus, camera calibration is considered important in obtaining high-accuracy UAV mapping using low-cost digital cameras. The main focus of this study was to calibrate the UAV camera at different camera distances and check the measurement accuracy. The scope of this study included camera calibration in the laboratory and on the field, and the UAV image mapping accuracy assessment used calibration parameters of different camera distances. The camera distances used for the image calibration acquisition and mapping accuracy assessment were 1.5 metres in the laboratory, and 15 and 25 metres on the field using a Sony NEX6 digital camera. A large calibration field and a portable calibration frame were used as the tools for the camera calibration and for checking the accuracy of the measurement at different camera distances. Bundle adjustment concept was applied in Australis software to perform the camera calibration and accuracy assessment. The results showed that the camera distance at 25 metres is the optimum object distance as this is the best accuracy obtained from the laboratory as well as outdoor mapping. In conclusion, the camera calibration at several camera distances should be applied to acquire better accuracy in mapping and the best camera parameter for the UAV image mapping should be selected for highly accurate mapping measurement.

Pseudo multi-exposure fusion using a single image

Abstract: This paper proposes a novel pseudo multi-exposure image fusion based on a single image. Multi-exposure image fusion is a method to produce images without saturation regions, by using photos with different exposures. However, it is difficult to take photos suited for the multi-exposure image fusion when we take a photo of dynamic scenes or record a video. In addition, the multi-exposure image fusion cannot be applied to existing images with a single exposure or videos. The proposed method enables us to produce pseudo multi-exposure images from a single image. To produce multi-exposure images, the proposed method utilizes a relationship between exposure values and pixel values. The relationship is obtained by assuming that a digital camera has a linear response function. Moreover, it is shown that the use of a local contrast enhancement method allows us to produce pseudo multi-exposure images with higher quality. Most of conventional multi-exposure image fusion methods are also applicable to the pseudo multi-exposure images. Experimental results show the effectiveness of the proposed method by comparing the proposed one with conventional ones.

Image data segmentation using depth data

Abstract: Devices and techniques are generally described for segmentation of image data using depth data. In various examples, color image data may be received from a digital camera. In some examples, depth image data may be received from a depth sensor. In various examples, the depth image data may be separated into a plurality of clusters of depth image data, wherein each cluster is associated with a respective range of depth values. In some further examples, a determination may be made that a first cluster of image data corresponds to an object of interest, such as a human subject, in the image data. In various examples, pixels of the first cluster may be encoded with foreground indicator data. In some further examples, segmented image data may be generated. The segmented image data may comprise pixels encoded with the foreground indicator data.

Unsupervised Learning for Color Constancy

Abstract: Most digital camera pipelines use color constancy methods to reduce the influence of illumination and camera sensor on the colors of scene objects. The highest accuracy of color correction is obtained with learning-based color constancy methods, but they require a significant amount of calibrated training images with known ground-truth illumination. Such calibration is time consuming, preferably done for each sensor individually, and therefore a major bottleneck in acquiring high color constancy accuracy. Statistics-based methods do not require calibrated training images, but they are less accurate. In this paper an unsupervised learning-based method is proposed that learns its parameter values after approximating the unknown ground-truth illumination of the training images, thus avoiding calibration. In terms of accuracy the proposed method outperforms all statistics-based and many learning-based methods. An extension of the method is also proposed, which learns the needed parameters from non-calibrated images taken with one sensors and which can then be successfully applied to images taken with another sensor. This effectively enables inter-camera unsupervised learning for color constancy. Additionally, a new high quality color constancy benchmark dataset with 1365 calibrated images is created, used for testing, and made publicly available. The results are presented and discussed. The source code and the dataset are available at this http URL

Underwater Image Systems Simulation

Abstract: We use modern computer graphics tools such as ray-tracing, digital camera simulation tools, and a physically accurate model of seawater constituents to simulate how light is captured by the imaging sensor in a digital camera placed in underwater ocean environments. Our water model includes parameters for the type and amount of phytoplankton, the concentration of chlorophyll, the amount of dissolved organic matter (CDOM) and the concentration of detritus (non-algal particles, NAP). We show that by adjusting the model parameters, we can predict real sensor data captured by a digital camera at a fixed distance and depth from a reference target with known spectral reflectance. We also provide an open-source implementation of all our tools and simulations.

Bus-based information management system

Abstract: The invention discloses a bus-based information management system, which belongs to the technical field of intelligent traffic. Through integrating multiple modern technical means such as a high-sensitivity alcohol detection sensor, a high-sensitivity gasoline detection sensor, a compartment temperature sensor, a smoke sensor, a standard definition wireless camera, a high-definition digital camera, a one-key alarm device, a sound-light alarm lamp, an infrared sensor, a Kinect depth camera, a mini PC, a 4G wireless router and an engine temperature sensor in a vehicle-mounted terminal, functions of video monitoring, identity recognition, passenger flow statistics, dangerous goods detection and linkage alarming are integrated as a whole. Thus, according to the bus-based information management system disclosed by the invention, multiple modern IT means are integrated, the global bus scheduling optimization and safety protection are enhanced, and the integrated management specifications of a public transportation system can be improved.

NDVI image extraction of an agricultural land using an autonomous quadcopter with a filter-modified camera

Abstract: The Normalized Difference Vegetation Index (NDVI) has been used in applications related to monitoring crops in agricultural areas This metric was used with automation to survey agricultural fields, and to provide an estimation of the conditions of crops in contrast to actual observations and care done by local farmers Having several agricultural areas in the country, this can be beneficial A quadcopter was used as the platform, equipped with a flight controller and a Raspberry Pi Zero that communicated with a Robot Operating System (ROS) for commands and data acquisition Through ROS, the quadcopter was further equipped with a filter-modified digital camera, and a GPS module Images taken by the camera were transferred to a computer for offline processing of the stitching and NDVI extraction of the images Stitching was done with Speeded Up Robust Features (SURF) and Scale Invariant Feature Transform (SIFT) NDVI was acquired using the blue band of the images containing near infrared (NIR) light data, and the red band containing visible light data The results showed that SIFT is more suitable for the study where image features had varying lighting and rotation; the values acquired from the final NDVI image showed consistency with the actual state of the corn crops observed in the study