Merge pull request #629 from Avdhesh-Varshney/bodyParts

KWOC 2023 : Body Parts Recognition

Body Parts Classification/Dataset/README.md

@@ -0,0 +1,20 @@
+# About the dataset
+Dataset can be downloaded from [here](https://www.kaggle.com/datasets/linkanjarad/body-parts-dataset)
+
+# Attributes in the dataset
+
+There are 10 directories in the dataset of different types of images.
+
+Images of:
+
+1) Belly
+2) Ear
+3) Elbow
+4) Eye
+5) Foot
+6) Hand
+7) Knee
+8) Neck
+9) Nose
+10) Shoulders
+

Body Parts Classification/Model/README.md

@@ -0,0 +1,36 @@
+# Body Parts Recognition Model
+
+# GOAL 
+The main goal of the project is to analyze the parts of the body and recognize them. On the basis of it, model is trained, to predict the body parts by input the pictures.
+
+# DATASET 
+* (Dataset Link)[https://www.kaggle.com/datasets/linkanjarad/body-parts-dataset]
+
+# WHAT I HAD DONE 
+* Data Exploration done with the help of EDA, to better understand the Body parts dataset and get useful insights.
+* I prepare dataset for model training and model testing purpose.
+* Then, normalizing the images dataset for 255 pixels color ration.
+* Now, build a neural network to learn from the prepared dataset.
+* At the end, adding a fully connected layers which do last step processing and recognizes the images.
+* Hence, did the Model Creation.
+
+# MODELS USED 
+* Sequential
+* Dense layers
+* Fully connected layers
+
+# LIBRARIES NEEDED 
+* cv2
+* tensorflow
+* keras
+* matplotlib
+
+# CONCLUSION 
+- By classification dataset into training and testing dataset figure is around 1669 and 429 images of the body parts.
+- By Using Neural network, We can get the minimum Loss value possible.
+- Here, deep neural network model can predict most accurate results for predicting body parts which is highest model performance in comparison with other models.
+- The Accuracy achieved by the model during training is around 99.75%.
+- The Accuracy achieved by the model during testing is around 91.88%.
+
+### Accuracy:- 91.88 %
+

Body Parts Classification/requirements.txt

@@ -0,0 +1,7 @@
+keras==2.15.0
+matplotlib==3.7.1
+numpy==1.23.5
+opencv-python==4.8.0.76
+pandas==1.5.3
+requests==2.31.0
+tensorflow==2.15.0