update project

artifacts/models/baseline_history.json

@@ -0,0 +1,154 @@
+{
+    "accuracy": [
+        0.8880633115768433,
+        0.9394962787628174,
+        0.9497249722480774,
+        0.9571552872657776,
+        0.9619801044464111,
+        0.9652610421180725,
+        0.9676734805107117,
+        0.9698928594589233,
+        0.9725947976112366,
+        0.9738492965698242,
+        0.9725947976112366,
+        0.9744282364845276,
+        0.9767441749572754,
+        0.976647675037384,
+        0.9782881140708923,
+        0.9805075526237488,
+        0.97954261302948,
+        0.9807970523834229,
+        0.9814725518226624,
+        0.983209490776062,
+        0.9830164909362793,
+        0.9819550514221191,
+        0.9837884902954102,
+        0.9834989905357361,
+        0.9827269911766052,
+        0.9844639301300049,
+        0.9837884902954102,
+        0.9862009286880493,
+        0.9859114289283752,
+        0.985042929649353,
+        0.9880343675613403,
+        0.9875518679618835,
+        0.9875518679618835,
+        0.9874553680419922,
+        0.9886133074760437,
+        0.9862009286880493
+    ],
+    "loss": [
+        0.2664412558078766,
+        0.15920524299144745,
+        0.13362552225589752,
+        0.11977039277553558,
+        0.10575573146343231,
+        0.09759275615215302,
+        0.09332476556301117,
+        0.08601260185241699,
+        0.08296018838882446,
+        0.078891821205616,
+        0.07702494412660599,
+        0.07292535156011581,
+        0.06834259629249573,
+        0.06703294813632965,
+        0.06486733257770538,
+        0.06227454915642738,
+        0.061870187520980835,
+        0.05987834930419922,
+        0.056517522782087326,
+        0.055569957941770554,
+        0.053644366562366486,
+        0.054151490330696106,
+        0.05065951496362686,
+        0.05058245360851288,
+        0.050053149461746216,
+        0.04905802756547928,
+        0.048753850162029266,
+        0.04615778475999832,
+        0.04582696408033371,
+        0.04523399844765663,
+        0.04218636080622673,
+        0.043016765266656876,
+        0.041500113904476166,
+        0.04166035354137421,
+        0.040624041110277176,
+        0.04107125103473663
+    ],
+    "val_accuracy": [
+        0.9391827583312988,
+        0.9502692222595215,
+        0.9550206065177917,
+        0.9543870687484741,
+        0.9645232558250427,
+        0.9623059630393982,
+        0.9680076241493225,
+        0.9695913791656494,
+        0.9680076241493225,
+        0.972758948802948,
+        0.9759265184402466,
+        0.9746595025062561,
+        0.9787773489952087,
+        0.9787773489952087,
+        0.9794108271598816,
+        0.9794108271598816,
+        0.9806778430938721,
+        0.9778270721435547,
+        0.9835286736488342,
+        0.9816281199455261,
+        0.9832119345664978,
+        0.9819448590278625,
+        0.9819448590278625,
+        0.9851124286651611,
+        0.9771935343742371,
+        0.9822616577148438,
+        0.9832119345664978,
+        0.9857459664344788,
+        0.9841621518135071,
+        0.9844789505004883,
+        0.9863794445991516,
+        0.9835286736488342,
+        0.9844789505004883,
+        0.9860627055168152,
+        0.9828951358795166,
+        0.9809946417808533
+    ],
+    "val_loss": [
+        0.17172108590602875,
+        0.13897046446800232,
+        0.12103567272424698,
+        0.11579054594039917,
+        0.10252440720796585,
+        0.10001140832901001,
+        0.08984819054603577,
+        0.08655385673046112,
+        0.08577841520309448,
+        0.07951025664806366,
+        0.07641823589801788,
+        0.07423911988735199,
+        0.07199306041002274,
+        0.07009323686361313,
+        0.06831575930118561,
+        0.06736170500516891,
+        0.06620226800441742,
+        0.06701282411813736,
+        0.06184401735663414,
+        0.06242649629712105,
+        0.06026426702737808,
+        0.05989912897348404,
+        0.06237339600920677,
+        0.05732772499322891,
+        0.06300686299800873,
+        0.059069763869047165,
+        0.05765041336417198,
+        0.05478701367974281,
+        0.05648797005414963,
+        0.055511001497507095,
+        0.0535995289683342,
+        0.0559844933450222,
+        0.05646961182355881,
+        0.053657740354537964,
+        0.058711983263492584,
+        0.05925580859184265
+    ]
+}

artifacts/models/category_history.json

@@ -0,0 +1,117 @@
+{
+    "accuracy": [
+        0.8836225271224976,
+        0.9570059776306152,
+        0.9707329869270325,
+        0.9783303141593933,
+        0.9834240078926086,
+        0.9850643277168274,
+        0.9886903166770935,
+        0.9917983412742615,
+        0.9954243302345276,
+        0.9938703179359436,
+        0.9939566850662231,
+        0.995078980922699,
+        0.9947336316108704,
+        0.9940429925918579,
+        0.9933523535728455,
+        0.9969783425331116,
+        0.9955106377601624,
+        0.996891975402832,
+        0.9987049698829651,
+        0.9983596801757812,
+        0.9986186623573303
+    ],
+    "loss": [
+        0.37353599071502686,
+        0.14305350184440613,
+        0.093997523188591,
+        0.06734441965818405,
+        0.049412257969379425,
+        0.04349873214960098,
+        0.03429267182946205,
+        0.02595502696931362,
+        0.017324281856417656,
+        0.017613478004932404,
+        0.018058663234114647,
+        0.016504919156432152,
+        0.01680760085582733,
+        0.01652861014008522,
+        0.019517360255122185,
+        0.010544461198151112,
+        0.01264826487749815,
+        0.009982614777982235,
+        0.005640930961817503,
+        0.0056995246559381485,
+        0.005192732438445091
+    ],
+    "val_accuracy": [
+        0.9691011309623718,
+        0.9846585988998413,
+        0.9889801144599915,
+        0.9950302243232727,
+        0.9967588782310486,
+        0.9958945512771606,
+        0.9924373626708984,
+        0.9987035393714905,
+        0.9982714056968689,
+        0.9989196062088013,
+        0.9969749450683594,
+        0.9984874725341797,
+        0.9989196062088013,
+        0.9984874725341797,
+        0.9978392124176025,
+        0.9997839331626892,
+        0.9987035393714905,
+        0.9991356730461121,
+        0.9991356730461121,
+        0.9993517994880676,
+        0.9993517994880676
+    ],
+    "val_loss": [
+        0.1030760183930397,
+        0.058829668909311295,
+        0.0396256297826767,
+        0.024300476536154747,
+        0.0160489771515131,
+        0.016302524134516716,
+        0.026506498456001282,
+        0.009227240458130836,
+        0.007496202364563942,
+        0.00561791704967618,
+        0.008364077657461166,
+        0.005430787336081266,
+        0.005307900719344616,
+        0.00564191909506917,
+        0.007319050841033459,
+        0.0017467719735577703,
+        0.004918231628835201,
+        0.0028124600648880005,
+        0.0021249386481940746,
+        0.002299233339726925,
+        0.0022428531665354967
+    ],
+    "learning_rate": [
+        0.0010000000474974513,
+        0.0010000000474974513,
+        0.0010000000474974513,
+        0.0010000000474974513,
+        0.0010000000474974513,
+        0.0010000000474974513,
+        0.0010000000474974513,
+        0.000800000037997961,
+        0.000800000037997961,
+        0.000800000037997961,
+        0.000800000037997961,
+        0.000800000037997961,
+        0.000800000037997961,
+        0.000800000037997961,
+        0.000800000037997961,
+        0.000640000042039901,
+        0.000640000042039901,
+        0.000640000042039901,
+        0.0005120000569149852,
+        0.0005120000569149852,
+        0.00040960003389045596
+    ]
+}

artifacts/models/category_labels.json

@@ -0,0 +1,12 @@
+[
+    "Apple",
+    "Banana",
+    "Bellpepper",
+    "Carrot",
+    "Cucumber",
+    "Mango",
+    "Orange",
+    "Potato",
+    "Strawberry",
+    "Tomato"
+]

requirements.txt

@@ -0,0 +1,13 @@
+numpy
+pandas
+matplotlib
+seaborn 
+tensorflow
+opencv-python
+fastapi
+uvicorn
+python-dotenv
+Pillow
+scikit-learn
+ultralytics 
+gradio

src/app/app.py

@@ -0,0 +1,41 @@
+import gradio as gr
+from src.pipeline.prediction_pipeline import PredictionPipeline
+import numpy as np
+from PIL import Image
+
+pipeline = PredictionPipeline()
+
+def predict_single(image):
+    
+    if image is None:
+        return None, "No image detected!", "No image detected!"
+    img = Image.fromarray(image) if isinstance(image, np.ndarray) else image
+    result = pipeline.predict(img)
+    annotated_img = pipeline.annotate(img, result)
+    return annotated_img, result["category"], result["freshness"]
+
+with gr.Blocks() as demo:
+    gr.Markdown("# Food Freshness Detection")
+
+    with gr.Tab("Image Upload"):
+        image = gr.Image(sources=["upload"], label="Upload an Image")
+        out_img = gr.Image()
+        cat = gr.Textbox(label="Category")
+        fresh = gr.Textbox(label="Freshness")
+        btn = gr.Button("Predict on Image")
+        btn.click(predict_single, inputs=image, outputs=[out_img, cat, fresh])
+
+        
+    with gr.Tab("Live Webcam"):
+        webcam = gr.Image(sources=["webcam"], label="Webcam")
+        out_img = gr.Image()
+        cat = gr.Textbox(label="Category")
+        fresh = gr.Textbox(label="Freshness")
+        btn = gr.Button("Predict")
+        btn.click(predict_single, inputs=webcam, outputs=[out_img, cat, fresh])
+
+
+    
+
+if __name__ == "__main__":
+    demo.launch(server_name="0.0.0.0", server_port=7860)

src/config.py

@@ -0,0 +1,18 @@
+import os 
+
+
+class Config:
+    PROJECT_ROOT = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
+
+    DATA_DIR = os.path.join(PROJECT_ROOT, 'artifacts', 'data')
+    RAW_DATA_DIR = os.path.join(PROJECT_ROOT, 'artifacts', 'data', 'raw')
+    PROCESSED_DATA_DIR = os.path.join(PROJECT_ROOT, 'artifacts', 'data', 'processed')
+    MODEL_DIR = os.path.join(PROJECT_ROOT, 'artifacts', 'models')
+    RESULTS_DIR = os.path.join(PROJECT_ROOT, 'artifacts', 'results')
+
+    METADATA_FILE = os.path.join(PROJECT_ROOT, 'artifacts', 'data', 'metadata.json')
+    CATEGORY_NAMES =  ['Fruits','Vegetables']
+    FRUIT_NAMES = ['Apple', 'Banana', 'Mango', 'Orange', 'Strawberry']
+    VEGETABLE_NAMES = ['Bellpepper', 'Carrot', 'Cucumber', 'Potato', 'Tomato' ]
+    CLASS_NAMES = ['Fresh', 'Rotten']
+  

src/pipeline/prediction_pipeline.py

@@ -0,0 +1,93 @@
+import os
+from pathlib import Path
+import numpy as np
+from PIL import Image, ImageDraw, ImageFont
+import tensorflow as tf
+import json
+
+from src.config import Config
+
+CATEGORY_MODEL_PATH = Path(Config.MODEL_DIR) / "category_classifier.keras"
+FRESHNESS_MODEL_PATH = Path(Config.MODEL_DIR) / "mobilenetv2_baseline.keras"
+
+CATEGORY_LABELS_PATH = Path(Config.MODEL_DIR) / "category_labels.json"
+with open(CATEGORY_LABELS_PATH, "r") as f:
+    CATEGORY_LABELS = json.load(f)
+
+FRESHNESS_LABELS = ["Fresh", "Rotten"]
+
+CATEGORY_IMG_SIZE = (224, 224)
+FRESHNESS_IMG_SIZE = (224, 224)
+
+class PredictionPipeline:
+    def __init__(self):
+        self.category_model = tf.keras.models.load_model(CATEGORY_MODEL_PATH)
+        self.freshness_model = tf.keras.models.load_model(FRESHNESS_MODEL_PATH)
+        self.category_labels = CATEGORY_LABELS
+        self.freshness_labels = FRESHNESS_LABELS
+
+    def _preprocess_image(self, img, target_size, normalize=True):
+        if isinstance(img, (str, Path)):
+            img_path = Path(img)
+            if not img_path.exists():
+                raise FileNotFoundError(f"Image not found: {img_path.resolve()}")
+            img = Image.open(str(img_path)).convert("RGB")
+        elif isinstance(img, np.ndarray):
+            if img.ndim == 3 and img.shape[2] == 3:
+                img = Image.fromarray(img)
+            else:
+                raise ValueError("NumPy input must be shape (H, W, 3)")
+        img_resized = img.resize(target_size)
+        img_array = np.array(img_resized).astype("float32")
+        if normalize:
+            img_array = img_array / 255.0
+        img_array = np.expand_dims(img_array, axis=0)
+        return img_array, img
+
+ 
+    def predict(self, img):
+   
+        cat_img_array, pil_img = self._preprocess_image(img, CATEGORY_IMG_SIZE, normalize=True)
+        cat_pred = self.category_model.predict(cat_img_array)
+        cat_idx = int(np.argmax(cat_pred))
+        cat_label = self.category_labels[cat_idx]
+        cat_score = float(np.max(cat_pred))
+
+      
+        fresh_img_array, _ = self._preprocess_image(img, FRESHNESS_IMG_SIZE, normalize=True)
+        fresh_pred = self.freshness_model.predict(fresh_img_array)
+        fresh_idx = int(np.argmax(fresh_pred))
+        fresh_label = self.freshness_labels[fresh_idx]
+        fresh_score = float(np.max(fresh_pred))
+
+        return {
+            "category": {"label": cat_label, "idx": cat_idx, "score": cat_score},
+            "freshness": {"label": fresh_label, "idx": fresh_idx, "score": fresh_score},
+            "pil_img": pil_img
+        }
+
+    def annotate(self, img, result, font_size=28):
+        pil_img = result["pil_img"]
+        draw = ImageDraw.Draw(pil_img)
+        text = f"{result['category']['label']} ({result['category']['score']:.2f}) | " \
+               f"{result['freshness']['label']} ({result['freshness']['score']:.2f})"
+        try:
+            font = ImageFont.truetype("arial.ttf", font_size)
+        except:
+            font = ImageFont.load_default()
+        draw.rectangle([0, 0, pil_img.width, font_size+8], fill=(0,0,0,160))
+        draw.text((5, 2), text, fill=(255, 255, 255), font=font)
+        return pil_img
+
+if __name__ == "__main__":
+    pipeline = PredictionPipeline()
+
+    img_path = r"artifacts\data\category\test\Apple\rottenApple (175).jpg"
+    
+    result = pipeline.predict(img_path)
+    
+    print("Prediction:", result["category"], "|", result["freshness"])
+
+    annotated_img = pipeline.annotate(img_path, result)
+
+    annotated_img.save(r"artifacts\results\annotated_result.jpg")