Upload 17 files

.gitattributes

@@ -33,3 +33,11 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+data/16_3450.png filter=lfs diff=lfs merge=lfs -text
+data/16_3675.png filter=lfs diff=lfs merge=lfs -text
+data/16_3735.png filter=lfs diff=lfs merge=lfs -text
+data/16_3900.png filter=lfs diff=lfs merge=lfs -text
+data/19_00350.png filter=lfs diff=lfs merge=lfs -text
+data/23_00961.png filter=lfs diff=lfs merge=lfs -text
+data/25_00272.png filter=lfs diff=lfs merge=lfs -text
+data/67_02661.png filter=lfs diff=lfs merge=lfs -text

Lib/Core.py

@@ -0,0 +1,87 @@
+import cv2
+import numpy as np
+import torch
+from matplotlib import pyplot as plt
+from ultralytics import YOLO
+from ultralytics.engine.results import Masks
+
+class CablePoleSegmentation():
+    def __init__(self, MODEL_PATH=None, retina_mask=False):
+        if not MODEL_PATH:
+            MODEL_PATH = "./weight/yolov8l-seg-pre100.onnx"
+        self._RetinaMask=retina_mask
+        self.Model = YOLO(MODEL_PATH)  # load a custom model
+
+
+    def RescaleTheMask(self, orijinal_image, masks):
+        _masks = []
+        for contour in masks:
+            b_mask = np.zeros(orijinal_image.shape[:2], np.uint8)
+            contour = contour.astype(np.int32)
+            contour = contour.reshape(-1, 1, 2)
+            mask = cv2.drawContours(b_mask, [contour], -1, (1, 1, 1), cv2.FILLED)
+            _masks += [mask]
+        return _masks
+
+
+    def Process(self, image):
+        with torch.no_grad():
+            results = self.Model(
+                image,
+                save=False,
+                show_boxes=False,
+                project="./result/",
+                conf=0.5,
+                retina_masks=self._RetinaMask,
+                stream=True
+            )
+
+        with torch.no_grad():
+            for result in results:
+                maskCountours = result.masks.xy
+                boxes = result.boxes.xyxy.int().cpu().numpy()
+                classes = result.boxes.cls.cpu().numpy()
+
+            rescaledMasks =  self.RescaleTheMask(image, maskCountours)
+            return rescaledMasks, boxes, classes, result.plot()
+
+
+
+    def PlotResults(self, masks, boxes, classes, original_image, result_image, mask, cable_mask):
+        fig, axs = plt.subplots(nrows=2, ncols=2, figsize=(27,15))
+        axs[0][0].imshow(original_image)
+        axs[0][0].set_title("Orijinal Görüntü")
+
+        axs[0][1].imshow(mask)
+        axs[0][1].set_title("Segmentasyon Maskesi")
+
+        
+        cv2.imwrite("cable_mask.png", cable_mask)
+        axs[1][0].imshow(cable_mask)
+        axs[1][0].set_title("Seçilen")
+
+        axs[1][1].imshow(result_image)
+        axs[1][1].set_title("Sonuç")
+        plt.show()
+
+
+
+if "__main__" == __name__:
+    test = "data/16_3450.png"
+    image = cv2.imread(test)
+    model = CablePoleSegmentation(retina_mask=True)
+    masks, boxes, classes, result_plot = model.Process(image)
+
+
+    fig, axs = plt.subplots(nrows=2, ncols=2, figsize=(27,15))
+    axs[0][0].imshow(image)
+    axs[0][0].set_title("Orijinal Görüntü")
+    axs[1][1].imshow(np.any(masks, axis=0))
+    axs[1][1].set_title("Sonuç")
+    plt.show()
+
+
+    # model.PlotResults(*model.Process(image))
+
+
+

Predict.py

@@ -0,0 +1,57 @@
+import cv2
+from matplotlib import pyplot as plt
+import numpy as np
+from ultralytics import YOLO
+import torch
+
+# Data
+test01 = "data/16_3450.png"
+test_image = test01
+
+# Load a model
+model = YOLO("weight/yolov8l-seg-pre100.onnx")  # load a custom model
+
+
+with torch.no_grad():
+    results = model(
+        test_image,
+        save=True,
+        show_boxes=False,
+        project="./result/",
+        conf=0.5,
+        retina_masks=False
+    )
+
+
+with torch.no_grad():
+    for result in results:
+        masks = result.masks.data
+        boxes = result.boxes.data
+
+    #ALL
+    canvas = torch.any(masks, dim=0).int() * 255
+
+    clss = boxes[:, 5]
+    obj_indices = torch.where(clss == 4)
+
+    # Cable
+    obj_masks = masks[obj_indices]
+    obj_mask = torch.any(obj_masks, dim=0).int() * 255
+    # cropped_image = result.orig_img[obj_mask.cpu().numpy()]
+
+
+fig, axs = plt.subplots(nrows=2, ncols=2, figsize=(27,15))
+axs[0][0].imshow(result.orig_img)
+axs[0][0].set_title("Orijinal Görüntü")
+
+axs[0][1].imshow(canvas.cpu().numpy())
+axs[0][1].set_title("Segmentasyon Maskesi")
+
+mask = np.array(obj_mask.cpu().numpy())*255
+cv2.imwrite("cable_mask.png", mask)
+axs[1][0].imshow(obj_mask.cpu().numpy())
+axs[1][0].set_title("Seçilen")
+
+axs[1][1].imshow(result.plot())
+axs[1][1].set_title("Sonuç")
+plt.show()

UI/Main.py

@@ -0,0 +1,77 @@
+import os
+import sys
+sys.path.append(os.getcwd())
+
+import numpy as np
+import gradio as gr
+import cv2
+from Lib.Core import CablePoleSegmentation
+
+
+demoImages = [
+    "data/16_3450.png", 
+    "data/16_3735.png",
+    "data/16_3900.png",
+    "data/19_00350.png",
+    "data/25_00272.png",
+    "data/67_02661.png"
+]
+labels = {0: "Boş", 1: "Çelik Direkler", 2: "Kablo", 3: "Kafes Kule", 4: "Kablo", 5: "Ahşap Kule"}
+color_map = {"Boş":"#ffffff", "Çelik Direkler":"#0000ff", "Kablo":"#00ff00", "Kafes Kule":"#ff0000", "Kablo":"#00ff00", "Ahşap Kule":"#ff0000"}
+MODEL = CablePoleSegmentation(retina_mask=False)
+
+def Warning():
+    gr.Info("DGH ARGE YAZILIM DANIŞMANLIK ENERJİ İNŞAAT SAN.TİC.LTD.ŞTİ", duration=0.5)
+
+with gr.Blocks(css="footer{display:none !important}") as block:
+    gr.Markdown("## Yüksek Gerilim Hattı Kablo ve Direk Tespit ve Segmentasyon Uygulaması - Demo")
+    
+
+    with gr.Row():
+        image = gr.Image(label="Fotoğraf")
+        processedImage = gr.Image(label="Sonuçlar")
+
+    with gr.Row() as imageGalleryRow:
+        imageGallery = gr.Gallery(
+            demoImages,
+            label="Demo Görsellerden Seçebilirsiniz",
+            rows=2, 
+            columns=4, 
+            object_fit="contain", 
+            height="auto"
+        )
+
+
+    results = gr.Textbox(label="Log")
+    processButton = gr.Button("Tespit Et")
+    annotatedImage = gr.AnnotatedImage(color_map=color_map)
+
+    @processButton.click(outputs=[processedImage, annotatedImage, results], inputs=image)
+    def Process(image):
+        MODEL = CablePoleSegmentation()
+        masks, boxes, classes, result_plot = MODEL.Process(image)
+
+        sections = []
+        
+        for m, b, c in zip(masks, boxes, classes):
+            sections+=[(m, labels[c])]
+            # sections+=[([*b], labels[c])]
+
+        return result_plot, (image, sections), "Görüntü İşlendi!"
+    
+
+    @imageGallery.select(inputs=imageGallery, outputs=[processedImage, annotatedImage, results])
+    def GalleryProcess(image, evt_data: gr.EventData):
+        selectedIdx = evt_data._data["index"]
+        imagePath = demoImages[selectedIdx]
+        image = cv2.imread(imagePath)
+        return Process(image)
+
+    
+    block.load(Warning)
+
+
+block.queue(max_size=10)
+block.launch(server_port=1071)
+
+

readme.txt

@@ -0,0 +1,4 @@
+1-Anaconda ortamı kurulur.
+
+2-Requirements içerisindeki kütüphaneler yüklenir.
+    $ conda install --yes --file requirements.txt -c pytorch -c nvidia

requirements.txt

@@ -0,0 +1,5 @@
+ultralytics
+pytorch
+torchvision
+torchaudio
+pytorch-cuda=12.4 

weight/yolov8l-seg-pre100.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bc705770eb7a2ac69bbc18fa811e48865c56426d9a7e413f7c3024eda3cabe25
+size 183915900