Browsing by Author "Kallepalli, Akhil"

Now showing 1 - 10 of 10
  • Thumbnail Image
    ItemOpen Access
    Application of spectral and spatial indices for specific class identification in Airborne Prism EXperiment (APEX) imaging spectrometer data for improved land cover classification
    (SPIE, 2016-09) Kallepalli, Akhil; Kumar, Arvind; Khoshelham, K.; James, David B.
    Hyperspectral remote sensing's ability to capture spectral information of targets in very narrow bandwidths gives rise to many intrinsic applications. However, the major limiting disadvantage to its applicability is its dimensionality, known as the Hughes Phenomenon. Traditional classification and image processing approaches fail to process data along many contiguous bands due to inadequate training samples. Another challenge of successful classification is to deal with the real world scenario of mixed pixels i.e. presence of more than one class within a single pixel. An attempt has been made to deal with the problems of dimensionality and mixed pixels, with an objective to improve the accuracy of class identification. In this paper, we discuss the application of indices to cope with the disadvantage of the dimensionality of the Airborne Prism EXperiment (APEX) hyperspectral Open Science Dataset (OSD) and to improve the classification accuracy using the Possibilistic c–Means (PCM) algorithm. This was used for the formulation of spectral and spatial indices to describe the information in the dataset in a lesser dimensionality. This reduced dimensionality is used for classification, attempting to improve the accuracy of determination of specific classes. Spectral indices are compiled from the spectral signatures of the target and spatial indices have been defined using texture analysis over defined neighbourhoods. The classification of 20 classes of varying spatial distributions was considered in order to evaluate the applicability of spectral and spatial indices in the extraction of specific class information. The classification of the dataset was performed in two stages; spectral and a combination of spectral and spatial indices individually as input for the PCM classifier. In addition to the reduction of entropy, while considering a spectral-spatial indices approach, an overall classification accuracy of 80.50% was achieved, against 65% (spectral indices only) and 59.50% (optimally determined principal components
  • Thumbnail Image
    ItemOpen Access
    Assessment of Light Transport Through Human Anatomy
    (Cranfield University, 2018-11-15 11:49) Kallepalli, Akhil
    Poster presented at the 2018 Defence and Security Doctoral Symposium.Due to the non-contact and non-invasive nature, and ease of investigation, photonics and imaging have emerged as a preferred modality in a multitude of domains. Investigation of medical conditions and problems utilising these tools is a highly sought after domain of research and innovation due to the impact it would have in improving the quality of care for patients while reducing the workload of medical staff. In our research, we consider simulations for light transport and subsequently study experimental evidence from interactions with the light of different wavelengths. As the light interacts with a complex combination of skin and underlying anatomy, it is absorbed, reflected and scattered light in characteristic ways. Our primary interest lies in better understanding light transport through the human anatomy, which is a complex, multi-layered target. Specifically, we are investigating the interaction of infrared and red wavelengths in a non-contact and non-invasive method. We are also delving into multi-sensor and multi-platform information fusion for monitoring ‘health’, directed towards the point-of-care segments of health care.
  • Thumbnail Image
    ItemOpen Access
    Digital shoreline analysis system-based change detection along the highly eroding Krishna–Godavari delta front
    (SPIE, 2017-08-24) Kallepalli, Akhil; Kakani, N. R.; James, David B.; Richardson, Mark A.
    Coastal regions are highly vulnerable to rising sea levels due to global warming. Previous Intergovernmental Panel on Climate Change (2013) predictions of 26 to 82 cm global sea level rise are now considered conservative. Subsequent investigations predict much higher levels which would displace 10% of the world’s population living less than 10 m above sea level. Remote sensing and GIS technologies form the mainstay of models on coastal retreat and inundation to future sea-level rise. This study estimates the varying trends along the Krishna–Godavari (K–G) delta region. The rate of shoreline shift along the 330-km long K–G delta coast was estimated using satellite images between 1977 and 2008. With reference to a selected baseline from along an inland position, end point rate and net shoreline movement were calculated using a GIS-based digital shoreline analysis system. The results indicated a net loss of about 42.1    km 2 area during this 31-year period, which is in agreement with previous literature. Considering the nature of landforms and EPR, the future hazard line (or coastline) is predicted for the area; the predication indicates a net erosion of about 57.6    km 2 along the K–G delta coast by 2050 AD.
  • Thumbnail Image
    ItemOpen Access
    Implications of spectral and spatial features to improve the identification of specific classes
    (SPIE, 2019-01-14) Kallepalli, Akhil; Kumar, Anil; Khoshelham, Kourosh; James, David B.; Richardson, Mark A.
    Dimensionality is one of the greatest challenges when deciphering hyperspectral imaging data. Although the multiband nature of the data is beneficial, algorithms are faced with a high computational load and statistical incompatibility due to the insufficient number of training samples. This is a hurdle to downstream applications. The combination of dimensionality and the real-world scenario of mixed pixels makes the identification and classification of imaging data challenging. Here, we address the complications of dimensionality using specific spectral indices from band combinations and spatial indices from texture measures for classification to better identify the classes. We classified spectral and combined spatial–spectral data and calculated measures of accuracy and entropy. A reduction in entropy and an overall accuracy of 80.50% was achieved when using the spectral–spatial input, compared with 65% for the spectral indices alone and 59.50% for the optimally determined principal components.
  • No Thumbnail Available
    ItemOpen Access
    Monte Carlo simulation results for anaemia detection in the skin
    (Cranfield University, 2020-01-20 08:27) Kallepalli, Akhil; James, David
    This research was presented at SPIE Photonics West 2020 (February 2020) with the article titled: Quantification and influence of skin chromophores for remote detection of anemic conditions (https://doi.org/10.1117/12.2545784) Monte Carlo simulations were performed using Virtual Tissue Simulator (VTS) and Monte Carlo Command Line (MCCL) tools. The target was a two-layer model of the epidermis and dermis layers of the skin. The optical attenuation properties quantifying the absorption and scattering in these layers are set according to two parameters: - Amount of melanin in the epidermis - Amount of haemoglobin in the blood-perfused dermis The optical models are simulated at six wavelengths for 6 skin types (Fitzpatrick's scale) and two blood conditions, healthy and anaemic. This results in a total of 72 simulations The simulation calculates the optical interaction of 10 6 photons with the various combinations of optical properties. The resulting outputs show the absorption and fluence in the tissue model, and reflected and transmitted energy.
  • Thumbnail Image
    ItemOpen Access
    Monte Carlo simulation results for full finger models based on ultrasound image data
    (Cranfield University, 2021-06-02 16:35) Kallepalli, Akhil; James, David; Richardson, Mark
    Monte Carlo simulations were performed using Virtual Tissue Simulator (VTS) and Monte Carlo Command Line (MCCL) tools. The target was a tissue model specific to individual participants, including all anatomical components in a finger. The thickness and depth measurements were ascertained using an 18 MHz ultrasound probe. This implementation is proposed as a low-cost and novel method for patient-specific photo-therapy. The optical models of 12 participants are assessed at four wavelengths of interest. The optical interaction with the tissue layers is calculated with 10 6 photons. The absorption, reflection, fluence and transmission are measured and quantified in the experiments.
  • Thumbnail Image
    ItemOpen Access
    Multi-modal assessment of light transport through biological tissue
    (2020-02) Kallepalli, Akhil; James, David B.; Richardson, Mark A.
    The advent of biomedical optics and understanding of light transport through tissues has gathered enormous popularity since the works of Britton Chance, Steven Jacques and Valery Tuchin. Understanding light transport has allowed diagnoses, and in vivo and noncontact assessment of tissue. This thesis provides an interdisciplinary approach to using various modalities for optically interacting with biological tissue. The methods focus on lowcost, non-contact, non-invasive and/or simple methods to assess biological tissue. Ultrasound imaging, which is a key radiological imaging tool in today’s hospitals, is combined with powerful ray tracing tools to provide a quantitative assessment of tissue. The thesis discusses four individual studies, linked through either their methods or applications. Specifically, the biological tissues that light interacts within this thesis are skin and its layers, muscle, bone and blood. Skin-safe lasers are used in the studies to interact with participants through simulations and experiments. Through the course of this research, I investigate the optical compatibility of human skin with synthetic skin samples known as human skin equivalents (HSEs)1 . The result is a novel assessment combining tissue engineering and biomedical optics. Secondly, a simulated analysis of light interaction with a two-layer model was subsequently analysed in a study to look for anaemic blood condition markers in the reflectance and fluence of photons. This study resulted in a unique assessment of light transport through two-layer models. The models accommodate melanin and haemoglobin concentrations in the layers of the skin, thereby accounting for all skin types and healthy and anaemic blood perfusion in the dermal layer. In a third study, the understanding and consideration of the influence of melanin and haemoglobin in the skin layers are extended to developing full-finger models. The full-finger models are based on high-frequency ultrasound image data2 . The optical models were assessed in visible and near-infrared wavelengths using Monte Carlo simulations. This provides a method to assess tissue damage before treatments such as photodynamic therapy. Finally, an image processing exercise to identify and monitor vascular activity was undertaken in the fourth study of this thesis. The vascular activity was imaged and monitored using a simple transmission-based experimental strategy and off-theshelf equipment. Vascular activity analogous to heart rate was successfully monitored for the participants of the study, accounting for motion of the finger in a non-contact experimental and processing workflow
  • Thumbnail Image
    ItemOpen Access
    Optical investigation of three-dimensional human skin equivalents: a pilot study
    (Wiley-VCH Verlag, 2019-10-08) Kallepalli, Akhil; McCall, Blake; James, David B.; Junaid, Sarah; Halls, James; Richardson, Mark A.
    Human skin equivalents (HSEs) are three‐dimensional living models of human skin that are prepared in vitro by seeding cells onto an appropriate scaffold. They recreate the structure and biological behaviour of real skin, allowing the investigation of processes such as keratinocyte differentiation and interactions between the dermal and epidermal layers. However, for wider applications, their optical and mechanical properties should also replicate those of real skin. We therefore conducted a pilot study to investigate the optical properties of HSEs. We compared Monte Carlo simulations of (1) real human skin and (2) two‐layer optical models of HSEs with (3) experimental measurements of transmittance through HSE samples. The skin layers were described using a hybrid collection of optical attenuation coefficients. A linear relationship was observed between the simulations and experiments. For samples thinner than 0.5 mm, an exponential increase in detected power was observed due to fewer instances of absorption and scattering.
  • Thumbnail Image
    ItemOpen Access
    Progress towards recalibration of spectrographs
    (IEEE, 2019-03-07) Kallepalli, Akhil; Soori, Umair; James, David; Richardson, Mark A.
    The spectral resolution of a spectrograph depends on the input slit width, the diffraction grating grooves and the number of imaging sensor/detector pixels. Due to the proprietary nature of spectrograph designs, recalibration by end-users can be challenging. Most calibration procedures currently published are applicable to in-house instruments or spectrographs with access to the internal specifications. Narrowing the input slit improves the resolution but also reduces the throughput of the imaging system. We attempted to recalibrate an Offner-based spectrograph by using a larger detector plane (an imaging system with a larger sensor), to vary the distance along the focal plane; and by utilising lens optics. Basic experiments were conducted by varying the distance from the exit window and inserting a lens to magnify the spectrograph output onto the larger detector plane. We concluded that the calibration could not be achieved using simple optics within the scope of our experiments. This article addresses a gap in literature that does not present the research community with the unsuccessful steps that are not applicable to similar problem statements. The alternative would be to rely on reflective optics, but this approach may reduce portability.
  • Thumbnail Image
    ItemOpen Access
    Where's Oxygen!
    (Cranfield University, 2017-12-11 15:48) Kallepalli, Akhil
    3MT presented at the 2017 Defence and Security Doctoral Symposium.May it be diagnostic medicine or surgical assistance, oxygen perfusion determination is one of the primary indicators of healthy tissue. In this talk, I will be discussing the utility and applications of non-contact oxygen estimation.