Hierarchically grouped 2D local features applied to edge contour localisation

One of the most significant research topics in computer vision is object detection. Most of the reported object detection results localise the detected object within a bounding box, but do not explicitly label the edge contours of the object. Since object contours provide a fundamental diagnostic of object shape, some researchers have initiated work on linear contour feature representations for object detection and localisation. However, linear contour feature-based localisation is highly dependent on the performance of linear contour detection within natural images, and this can be perturbed significantly by a cluttered background. In addition, the conventional approach to achieving rotation-invariant features is to rotate the feature receptive field to align with the local dominant orientation before computing the feature representation. Grid resampling after rotation adds extra computational cost and increases the total time consumption for computing the feature descriptor. Though it is not an expensive process if using current computers, it is appreciated that if each step of the implementation is faster to compute especially when the number of local features is increasing and the application is implemented on resource limited ”smart devices”, such as mobile phones, in real-time. Motivated by the above issues, a 2D object localisation system is proposed in this thesis that matches features of edge contour points, which is an alternative method that takes advantage of the shape information for object localisation. This is inspired by edge contour points comprising the basic components of shape contours. In addition, edge point detection is usually simpler to achieve than linear edge contour detection. Therefore, the proposed localization system could avoid the need for linear contour detection and reduce the pathological disruption from the image background. Moreover, since natural images usually comprise many more edge contour points than interest points (i.e. corner points), we also propose new methods to generate rotation-invariant local feature descriptors without pre-rotating the feature receptive field to improve the computational efficiency of the whole system. In detail, the 2D object localisation system is achieved by matching edge contour points features in a constrained search area based on the initial pose-estimate produced by a prior object detection process. The local feature descriptor obtains rotation invariance by making use of rotational symmetry of the hexagonal structure. Therefore, a set of local feature descriptors is proposed based on the hierarchically hexagonal grouping structure. Ultimately, the 2D object localisation system achieves a very promising performance based on matching the proposed features of edge contour points with the mean correct labelling rate of the edge contour points 0.8654 and the mean false labelling rate 0.0314 applied on the data from Amsterdam Library of Object Images (ALOI). Furthermore, the proposed descriptors are evaluated by comparing to the state-of-the-art descriptors and achieve competitive performances in terms of pose estimate with around half-pixel pose error.

An examination of verbal descriptors of cancer-induced bone pain and neuropathic cancer pain

Aims: The aim of the thesis was to identify verbal descriptors of cancer induced bone pain (CIBP) and neuropathic cancer pain (NCP). An examination of the verbal descriptors associated with these two pain syndromes further considered the relationship between common verbal descriptors, cancer type, performance status and analgesia. Methods: The project was conducted in two phases; Phase one was a systematic review of the literature to examine current evidence of verbal descriptors in CIBP and NCP. Phase two utilised secondary data analysis methodology. Data from 120 patients with confirmed CIBP and 61 patients with confirmed NCP were deemed eligible for entry into a de novo database for secondary analysis. Key descriptive data were considered such as gender, ECOG and pain scores to characterise the patient population. Verbal descriptors of CIBP and NCP were considered in detail across the secondary de novo database. Results: Gender was not identified as a diagnostic characteristic of CIBP and NCP with similar distribution across prevalence of pain reporting and also pain severity. Patients with breast (n=52,43.3%), prostate (n=35,29.2%) and lung (n=14,11.7%) cancer were found to be at an increased risk of CIBP. Those with NCP more was found more commonly among patients with breast cancer (n=21,34.4%). Patients with CIBP were found to have an ECOG performance of 1 (n=49, 40.8%) or 2 (n=43, 35.8%) which was lower than those with NCP with an ECOG of 0 (n=32, 52.5%) or 2 (n=18, 29.5%). Comparisons were made across analgesia and treatment options for CIBP and NCP. Patients with CIBP received a greater variety of treatment options including bisphosphonates and radiotherapy while patients with NCP were more commonly treated with analgesia alone. Patients with CIBP and NCP were taking strong opioids, however those with NCP (n=45, 73.8%) were more likely to utilise strong opioids than those with CIBP (n=61, 50.8%). It was noted that those with NCP required a daily morphine equivalence of almost 50% higher than those with CIBP. Average consumption of opioids was 155.6mg, for patients with NCP, compared to 76mg in patients with CIBP. Common verbal descriptors of CIBP and NCP were identified. The most common verbal descriptors for CIBP were aching, gnawing and throbbing and the most common verbal descriptors of NCP were aching, tender and sharp. Of the most common 6 descriptors for CIBP and NCP only one descriptor was unique to each pain type, gnawing for CIBP and stabbing for NCP. Conclusions: Patients with CIBP and NCP use similar verbal descriptors to characterise their pain with gnawing being unique to CIBP and stabbing being unique to NCP in the data considered within project. Further research is required to explore verbal descriptors which are both common and unique to CIBP and NCP. Further exploration of verbal descriptors would assist development of a comprehensive pain assessment tool which would enhance pain assessment for nurses, clinicians and patients.