Multistream video fusion using local principal components analysis

Sharma, Ravi K.; Pavel, Misha; Leen, Todd K.

doi:10.1117/12.327992

Cited by 11 publications

(4 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…PCA was also helpful in detecting the correlations between traits, as well as recognizing the independent principle components that are effective on plant attributes and assessing diversity [ 58 ]. Moreover, PCA is a well-known dimension reduction approach for condensing a big collection of variables into a small set that retains the majority of the information from the large set [ 59 ]. It reflects the importance of the largest contributor to the total variation at each axis of differentiation [ 60 ].…”

Section: Discussionmentioning

confidence: 99%

Multivariate analysis for yield and yield-related traits of amaranth genotypes from Ethiopia

Yeshitila¹,

Gedebo²,

Tesfaye³

et al. 2023

Heliyon

View full text Add to dashboard Cite

Section: Discussionmentioning

confidence: 99%

Multivariate analysis for yield and yield-related traits of amaranth genotypes from Ethiopia

Yeshitila¹,

Gedebo²,

Tesfaye³

et al. 2023

Heliyon

View full text Add to dashboard Cite

“…The combination of images provided by sensors operating at different bandwidths, such as thermal and visible, is a particularly challenging case for pixel-based fusion methods since the opposite contrast issue must be addressed in order to generate and to display the fused images. Techniques based on principal component analysis 4,15 , discrete wavelet transform 16,17,18 , opponent color contrast 7,19 or on Retinex algorithm 5 may be used for that purpose. For completely autonomous applications, the implementation of such methods may be avoided since there is no need to display new images that make human perception easier.…”

Section: Color / Thermal Image Fusionmentioning

confidence: 99%

<title>Context-independant video monitoring of mobile objects with color/thermal sensor</title>

2004

View full text Add to dashboard Cite

Automated video monitoring of mobile objects is a growing trend in many sectors, especially in surveillance applications. Many research groups are addressing substantial efforts to develop autonomous applications able to recognize specified events. Reliability of such a system is mainly defined by its ability to extract and to track features of interest in image sequences. Since the capacity to perform this basic task is strongly related to changes in image contrast, video monitoring units made up mono-spectral sensors offer limited performances in many situations. The best example of such a limitation is the uselessness of a visible CCD camera in low brightness scene. To overcome these shortcomings, we developed an acquisition unit including an uncooled VO x thermal camera (8-12 µm) and a HighDynamic-Range-CMOS (R) color camera more suitable for outdoor applications. Unlike similar systems, we perform image registration at hardware level rather than at software level. Advantageous characteristics of such a design are presented in this paper. A simple framework is also proposed in order to achieve context-independent event extraction from color and thermal information.

show abstract

“…Industry and academia efforts over the past few years have shown that combination of multi-spectral imagery can be either beneficial or detrimental, depending upon which source provides what information at which time [3][4][5][6][7][8][9][10][11][12]. In order to maximize benefit from the available information, a statistical relationship between the image sources must be derived such that the corresponding pixel amplitudes may be combined by some desired mathematical principal.…”

Section: Image Fusionmentioning

confidence: 99%

Processing system for an enhanced vision system

Yelton

Bernier

Sanders-Reed

2004

Enhanced and Synthetic Vision 2004

View full text Add to dashboard Cite

Enhanced Vision Systems (EVS) combines imagery from multiple sensors, possibly running at different frame rates and pixel counts, on to a display. In the case of a Helmet Mounted Display (HMD), the user line of sight is continuously changing with the result that the sensor pixels rendered on the display are changing in real time. In an EVS, the various sensors provide overlapping fields of view which requires stitching imagery together to provide a seamless mosaic to the user. Further, different modality sensors may be present requiring the fusion of imagery from the sensors. All of this takes place in a dynamic flight environment where the aircraft (with fixed mounted sensors) is changing position and orientation while the users are independently changing their lines of sight. In order to provide well registered, seamless imagery, very low throughput latencies are required, while dealing with huge volumes of data. This provides both algorithmic and processing challenges which must be overcome to provide a suitable system. This paper discusses system architecture, efficient stitching and fusing algorithms, and hardware implementation issues.

show abstract

Multistream video fusion using local principal components analysis

Cited by 11 publications

References 11 publications

Multivariate analysis for yield and yield-related traits of amaranth genotypes from Ethiopia

Multivariate analysis for yield and yield-related traits of amaranth genotypes from Ethiopia

<title>Context-independant video monitoring of mobile objects with color/thermal sensor</title>

Processing system for an enhanced vision system

Contact Info

Product

Resources

About