Significance of Haralick Features in Bone Tumor Classification Using Support Vector Machine

Distributed Mirror-Prox Optimization for Multi-Access Edge Computing

“…Let us denote the amount of offloaded data to the nth MEC server by the mth SD at time-slot t as a mn t (bits), the CPU-cycle frequency of the nth MEC server at time-slot t as ⇢ n t (cycles/sec), which cannot exceed its maximum value ⇢ u (cycles/sec). Let the small scale fading, e.g., [12], power gain between the mth SD and the BS be denoted by m t at time-slot t, which is exponentially distributed with mean 1 ⇠ , i.e., m t ⇠ Exp(⇠). Hence, the corresponding channel power gain can be expressed as h m t = m t q 0 (l 0 /l t ) ✓ , where q 0 is the path-loss constant, l 0 is the reference distance, l t is the distance between the mth SD and the BS in the tth time-slot, and ✓ is the path-loss exponent.…”

“…Therefore, to solve the problem in (12), we use projection gradient descent to choose the primal variable x t+1 followed by a projection gradient ascent to determine the corresponding dual variable t+1 as functions of the decisions made in timeslot t. Leveraging the idea of the prox-method [14], we define an auxiliary function A t (x, ), as…”

“…After the actions have been made, the cost function and constraint function will be revealed at the end of the time-slot. Then, we can update the auxiliary variables xt+1 , ˆ t+1 based on the dual problem in (12) as…”

A Sustainable Deep Learning Framework for Object Recognition Using Multi-Layers Deep Features Fusion and Selection

“…Although in the past few decades, the conventional approaches, such as Hand-Crafted Features (HCF), were used, as the time passed, however, the objects and their backgrounds became more confusing, thereby restricting their use. Handcrafted features included Histogram of Oriented Graph (HOG) [6], geometric features [7], Scale Invariant Feature Transformation (SIFT) [8], Difference of Gaussian (DoG) [9], Speeded-Up Robust Features (SURF) [10], and texture features (HARLICK) [11]. Recent techniques, in contrast, proposed to exploit a hybrid set of features to get a better representation of an object [12].…”