Chingri
Description
To describe the dataset of prawn images captured for the classification of "good" and "bad" prawns, here’s a *Dataset Overview:* This dataset comprises over 500 images of prawns, specifically categorized as either "good" or "bad." The images were captured using a Redmi Note 8 Pro mobile camera. The prawns were photographed on a black background under daylight conditions to ensure consistent lighting and minimize reflections or shadows. *Image Acquisition Setup:* - *Device:* Redmi Note 8 Pro mobile camera. - *Lighting Condition:* Daylight, ensuring natural and uniform illumination across all images. - *Background:* Plain black background, chosen to create high contrast between the prawn and the background, thereby enhancing the visibility of features that distinguish good from bad prawns. *Prawn Classification Criteria:* 1. *Good Prawns:* - *Physical Appearance:* Intact body structure with no visible damage or discoloration. - *Color:* A healthy, natural color that is typical for the species. - *Size and Shape:* Uniform size and shape that meets the standard for quality prawns. - *Texture:* Firm texture without any signs of mushiness or degradation. - *Smell (Inferred):* Fresh prawns typically have a clean, oceanic smell, which cannot be captured in images but is often inferred from physical appearance. 2. *Bad Prawns:* - *Physical Damage:* Visible signs of damage such as broken limbs, cracks, or deformities. - *Discoloration:* Any off-color areas, particularly browning or dark spots, indicating spoilage or disease. - *Size and Shape:* Irregular size or misshapen prawns that fall outside the desired specifications. - *Texture:* Appearance of a soft or mushy texture, often indicating spoilage. - *Additional Defects:* Presence of foreign matter, exoskeleton damage, or any other visual cues indicating poor quality. *Data Labeling:* Each image in the dataset has been labeled as either "good" or "bad" based on the above criteria. The labeling process involved both manual inspection and expert evaluation to ensure accuracy. *Dataset Usage:* This dataset is intended for use in machine learning models aimed at automating the classification of prawn quality. It can be used to train algorithms in distinguishing between good and bad prawns based on visual features, facilitating quality control processes in the seafood industry. *Image Details:* - *Resolution:* High-resolution images suitable for detailed analysis. - *File Format:* The images are stored in standard formats like JPEG or PNG, ensuring compatibility with various image processing tools. This description provides a detailed yet concise overview of the dataset, highlighting key aspects of the prawn images, classification criteria, and potential uses in research and industry.