Melanoma is the major cause for death worldwide also called malignant skin cancer. Early diagnosis of melanoma by dermatoscopy particularly improves survival. However, exact detection of melanoma is very difficult for following reasons. Low contrast among lesions and skin, visually similar among melanoma and non-melanoma lesions, etc. Reliable, automated diagnosis of skin tumors would therefore greatly help improve accuracy and efficacy of pathologists. Here, Binarized Spiking Neural Network optimized with Giza Pyramids Construction Optimization Algorithm for automatic Melanoma Classification (BSNNGPCOA-AMC-DI) from dermoscopic images is proposed. Primarily, the input Skin dermoscopic images are engaged from the dataset of Skin Lesion Images for Melanoma Classification. Then, the input Skin dermoscopic images is per-processed using Structural interval gradient filtering for removing noise and increase the quality of Skin dermoscopic images. Next, these pre-processed images are given to adaptive density-based spatial clustering for segmenting ROI region. The segmented ROI region is given into Ternary pattern and discrete wavelet transform for extracting Radiomic features such as Grayscale statistic features (standard deviation, mean, kurtosis, and skewness) and Haralick Texture features (contrast, energy, entropy, homogeneity, and inverse different moments). The extracted features are given into the Binarized Spiking Neural Networks which classifies the skin cancers such as Melanoma, nevus, Basal cell carcinoma, Actinic Keratosis, Pigmented Benign Keratosis, Dermatofibroma, Vascular lesion, Squamous cell carcinoma and seborrheic keratosis. In general, Binarized Spiking Neural Networks does not express any adaption of optimization strategies for determining the optimal parameters to assure accurate classification of skin cancer. Hence, Giza Pyramids Construction Optimization Algorithms proposed in this work to optimize the Binarized Spiking Neural Networks classifier, which precisely classifies the skin cancer. The proposed BSNN-GPCOA-AMC-DI method is implemented in MATLAB and the effectiveness is assessed with several performance metrics, like accuracy, precision, F score, sensitivity, specificity, ROC, computational time. Efficiency of proposed method BSNN - GPCOA-AMC-DI approach attains 10.12%, 22.33% high accuracy and 10.09%, 13.65% high precision is likened to the existing methods, such as Skin cancer classification of Convolutional Neural Network with optimized squeeze Net by Bald Eagle Search optimization (SCC - CNN - Squeeze Net - BES) and Skin cancer detection of Convolutional Neural Network using Gray Wolf Optimization (SCD - CNN - GWO) respectively.