{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"# Detecting Wheat Heads: Acquising Head Density Healt and Mateurity Information by Analysis of the Images\n \n -----","metadata":{}},{"cell_type":"markdown","source":"### ABSTRACT\n\nIn our study, the aim is to track wheat spikes and extract essential health information such as spike density, maturity, and basic health indicators from optical images to ensure a healthy growth process. To achieve this goal, image processing and machine learning methods are employed to obtain fundamental health data from wheat photos.\n\nThese data are intended to benefit wheat producers by enabling the development of solutions related to losses in the production process and the attainment of high-quality products.\n\n\n****  \nThis notebook describes the content of the \"IMAGE OPTIMIZATION\" and \"ANALYTIC APPROACH\" that is part of this work.\n\n![117530943-9c085f80-afe8-11eb-846c-af664f6b72ec.png](attachment:9cfe608f-550d-4020-8b5f-7a52b2eb88f3.png)","metadata":{},"attachments":{"9cfe608f-550d-4020-8b5f-7a52b2eb88f3.png":{"image/png":"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"}}},{"cell_type":"code","source":"# IMPORT LIBRARY\n\nimport os\nimport datetime\nimport numpy as np\nimport pandas as pd\nfrom matplotlib import pyplot as plt\nimport cv2\n\n\n\ndef printime(title):\n    print(\"{0}: {1}\".format(title, datetime.datetime.now()))","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","execution":{"iopub.status.busy":"2023-05-26T19:54:20.770313Z","iopub.execute_input":"2023-05-26T19:54:20.770821Z","iopub.status.idle":"2023-05-26T19:54:20.956648Z","shell.execute_reply.started":"2023-05-26T19:54:20.770779Z","shell.execute_reply":"2023-05-26T19:54:20.955319Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"We detected wheat ears using the EfficientDet object detection model we had trained before and the results were recorded on csv. Now we will load these csv files.","metadata":{}},{"cell_type":"code","source":"# Load Model Results\nlatesub = pd.read_csv(\"../input/wheat-latesub/submission.csv\")\nprint(\"Trained model summary output shape:\",latesub.shape)\nimageid_count = latesub['image_id'].value_counts()\nprint(\"Example dataset image count: \",len(imageid_count))\nprint(imageid_count)\nlatesub.head()","metadata":{"execution":{"iopub.status.busy":"2023-05-26T19:54:21.047678Z","iopub.execute_input":"2023-05-26T19:54:21.048065Z","iopub.status.idle":"2023-05-26T19:54:21.100418Z","shell.execute_reply.started":"2023-05-26T19:54:21.048033Z","shell.execute_reply":"2023-05-26T19:54:21.099533Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"latesub[\"PredictionString\"] = latesub[\"PredictionString\"].str.split(' ')\nlatesub.head()","metadata":{"execution":{"iopub.status.busy":"2023-05-26T19:54:21.726061Z","iopub.execute_input":"2023-05-26T19:54:21.726486Z","iopub.status.idle":"2023-05-26T19:54:21.741751Z","shell.execute_reply.started":"2023-05-26T19:54:21.726448Z","shell.execute_reply":"2023-05-26T19:54:21.740944Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Editting Uploaded Results\ndef preparingdf(df):\n    score = []\n    tx = []\n    ty = []\n    tw = []\n    th = []\n    for i in df.iterrows():\n        idx0 = i[1][1][0]\n        idx1 = i[1][1][1]\n        idx2 = i[1][1][2]\n        idx3 = i[1][1][3]\n        idx4 = i[1][1][4]\n        score.append(idx0)\n        tx.append(idx1)\n        ty.append(idx2)\n        tw.append(idx3)\n        th.append(idx4)\n        \n    df[['score']] = score  \n    df[['tx']] = tx  # bbox xmin\n    df[['ty']] = ty  # bbox ymin\n    df[['tw']] = tw  # bbox width\n    df[['th']] = th  # bbox heigth\n    \n\n    return df\n\n\ndata = preparingdf(latesub)\ndata.score = data.score.astype(float)\ndata.tx = data.tx.astype(float)\ndata.ty = data.ty.astype(float)\ndata.tw = data.tw.astype(float)\ndata.th = data.th.astype(float)\ndata.head()","metadata":{"execution":{"iopub.status.busy":"2023-05-26T19:54:46.454657Z","iopub.execute_input":"2023-05-26T19:54:46.455205Z","iopub.status.idle":"2023-05-26T19:54:46.533804Z","shell.execute_reply.started":"2023-05-26T19:54:46.455151Z","shell.execute_reply":"2023-05-26T19:54:46.532811Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data.groupby(['image_id'])['score'].max()","metadata":{"execution":{"iopub.status.busy":"2023-05-26T19:54:59.215406Z","iopub.execute_input":"2023-05-26T19:54:59.215825Z","iopub.status.idle":"2023-05-26T19:54:59.231017Z","shell.execute_reply.started":"2023-05-26T19:54:59.215791Z","shell.execute_reply":"2023-05-26T19:54:59.229661Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df = data.groupby(['image_id']).mean()\n\ndf[['count']] = imageid_count\ndf[['score_min']] = data.groupby(['image_id'])['score'].min()\ndf[['score_max']] = data.groupby(['image_id'])['score'].max()\ndf[['tw_min']] = data.groupby(['image_id'])['tw'].min()\ndf[['tw_max']] = data.groupby(['image_id'])['tw'].max()\ndf[['th_min']] = data.groupby(['image_id'])['th'].min()\ndf[['th_max']] = data.groupby(['image_id'])['th'].max()\n\ndf['FX'] = (1024*1024)//(df.tw_max*df.th_max)\ndf['FY'] = df['count']/df.FX\n\ndf.head()","metadata":{"execution":{"iopub.status.busy":"2023-05-26T19:55:50.114556Z","iopub.execute_input":"2023-05-26T19:55:50.115028Z","iopub.status.idle":"2023-05-26T19:55:50.199401Z","shell.execute_reply.started":"2023-05-26T19:55:50.114986Z","shell.execute_reply":"2023-05-26T19:55:50.198588Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"***Attention***  \nThere are 2 specially created functions here:\n1)  FX: *This function expresses the proportional value of the largest wheat ear in the whole image.*  \n2)  FY: *This function associates the FX function with the number of wheat.*","metadata":{}},{"cell_type":"code","source":"df.columns","metadata":{"execution":{"iopub.status.busy":"2023-05-26T19:56:08.349164Z","iopub.execute_input":"2023-05-26T19:56:08.349831Z","iopub.status.idle":"2023-05-26T19:56:08.358752Z","shell.execute_reply.started":"2023-05-26T19:56:08.349771Z","shell.execute_reply":"2023-05-26T19:56:08.357487Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df.shape","metadata":{"execution":{"iopub.status.busy":"2023-05-26T19:56:08.473298Z","iopub.execute_input":"2023-05-26T19:56:08.473715Z","iopub.status.idle":"2023-05-26T19:56:08.480978Z","shell.execute_reply.started":"2023-05-26T19:56:08.473674Z","shell.execute_reply":"2023-05-26T19:56:08.479678Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"columns = ['score', 'tx', 'ty', 'tw', 'th']\ndf.drop(columns, inplace=True, axis=1)\ndf.head()","metadata":{"execution":{"iopub.status.busy":"2023-05-26T19:56:25.493742Z","iopub.execute_input":"2023-05-26T19:56:25.494166Z","iopub.status.idle":"2023-05-26T19:56:25.519258Z","shell.execute_reply.started":"2023-05-26T19:56:25.494128Z","shell.execute_reply":"2023-05-26T19:56:25.518279Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"","metadata":{}},{"cell_type":"code","source":"# Preparing our sample test images...\ntest_dir = \"../input/global-wheat-detection/test/\"\ntestfile = os.listdir(test_dir)","metadata":{"execution":{"iopub.status.busy":"2023-05-26T19:57:54.551480Z","iopub.execute_input":"2023-05-26T19:57:54.551914Z","iopub.status.idle":"2023-05-26T19:57:54.559320Z","shell.execute_reply.started":"2023-05-26T19:57:54.551878Z","shell.execute_reply":"2023-05-26T19:57:54.558075Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"len(testfile)","metadata":{"execution":{"iopub.status.busy":"2023-05-26T19:57:57.113440Z","iopub.execute_input":"2023-05-26T19:57:57.113848Z","iopub.status.idle":"2023-05-26T19:57:57.119948Z","shell.execute_reply.started":"2023-05-26T19:57:57.113813Z","shell.execute_reply":"2023-05-26T19:57:57.118639Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Uploading test images\nimg = cv2.imread(test_dir+testfile[2])\nimg.shape","metadata":{"execution":{"iopub.status.busy":"2023-05-26T19:57:57.256105Z","iopub.execute_input":"2023-05-26T19:57:57.256535Z","iopub.status.idle":"2023-05-26T19:57:57.286080Z","shell.execute_reply.started":"2023-05-26T19:57:57.256501Z","shell.execute_reply":"2023-05-26T19:57:57.284791Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Converting the image from BGR format to LAB format\nlab = cv2.cvtColor(img, cv2.COLOR_BGR2LAB)\n\nl = lab[:,:,0]\na = lab[:,:,1]\nb = lab[:,:,2]","metadata":{"execution":{"iopub.status.busy":"2023-05-26T19:58:48.634444Z","iopub.execute_input":"2023-05-26T19:58:48.634887Z","iopub.status.idle":"2023-05-26T19:58:48.645726Z","shell.execute_reply.started":"2023-05-26T19:58:48.634847Z","shell.execute_reply":"2023-05-26T19:58:48.644373Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def summary(test_dir):\n    testfiles = os.listdir(test_dir)\n    for i in len(testfiles):\n        img = cv2.imread(test_dir+testfile[i])\n        lab = cv2.cvtColor(img, cv2.COLOR_BGR2LAB)\n        l = lab[:,:,0]\n        a = lab[:,:,1]\n        b = lab[:,:,2]\n        clahe = cv2.createCLAHE(clipLimit=2, tileGridSize=(16,16))\n        clahe_img = clahe.apply(l)\n        img_result = cv2.merge((clahe_img,a,b))\n        imgCLAHE = cv2.cvtColor(img_result, cv2.COLOR_LAB2RGB)","metadata":{"execution":{"iopub.status.busy":"2023-05-26T19:58:49.410249Z","iopub.execute_input":"2023-05-26T19:58:49.410666Z","iopub.status.idle":"2023-05-26T19:58:49.418718Z","shell.execute_reply.started":"2023-05-26T19:58:49.410629Z","shell.execute_reply":"2023-05-26T19:58:49.417701Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Histogram output of image in LAB format\nplt.hist(l.flat, bins=100, range=(0,255))","metadata":{"execution":{"iopub.status.busy":"2023-05-26T19:58:50.881150Z","iopub.execute_input":"2023-05-26T19:58:50.881542Z","iopub.status.idle":"2023-05-26T19:58:58.851350Z","shell.execute_reply.started":"2023-05-26T19:58:50.881504Z","shell.execute_reply":"2023-05-26T19:58:58.850449Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Applying the CHALE method to the above LAB format\nclahe = cv2.createCLAHE(clipLimit=2, tileGridSize=(16,16))\nclahe_img = clahe.apply(l)\n\nplt.hist(clahe_img.flat, bins=100, range=(0,255))","metadata":{"execution":{"iopub.status.busy":"2023-05-26T19:59:01.993540Z","iopub.execute_input":"2023-05-26T19:59:01.993950Z","iopub.status.idle":"2023-05-26T19:59:09.787289Z","shell.execute_reply.started":"2023-05-26T19:59:01.993917Z","shell.execute_reply":"2023-05-26T19:59:09.786039Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"img_result = cv2.merge((clahe_img,a,b))\nimgNORM = cv2.cvtColor(lab, cv2.COLOR_LAB2RGB) # CHALE uygulanmamış / not apply CHALE\nimgCLAHE = cv2.cvtColor(img_result, cv2.COLOR_LAB2RGB) # CHALE uygulanmış / apply CHALE","metadata":{"execution":{"iopub.status.busy":"2023-05-26T19:59:15.544459Z","iopub.execute_input":"2023-05-26T19:59:15.544917Z","iopub.status.idle":"2023-05-26T19:59:15.572777Z","shell.execute_reply.started":"2023-05-26T19:59:15.544871Z","shell.execute_reply":"2023-05-26T19:59:15.571706Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Görüntünün ilk hali... / First State\nplt.imshow(imgNORM)","metadata":{"execution":{"iopub.status.busy":"2023-05-26T19:59:23.363648Z","iopub.execute_input":"2023-05-26T19:59:23.364045Z","iopub.status.idle":"2023-05-26T19:59:23.662468Z","shell.execute_reply.started":"2023-05-26T19:59:23.364012Z","shell.execute_reply":"2023-05-26T19:59:23.660991Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# CHALE uygulandıktan sonra... / Last State with apply CHALE\nplt.imshow(imgCLAHE)","metadata":{"execution":{"iopub.status.busy":"2023-05-26T19:59:26.234330Z","iopub.execute_input":"2023-05-26T19:59:26.236871Z","iopub.status.idle":"2023-05-26T19:59:26.626607Z","shell.execute_reply.started":"2023-05-26T19:59:26.236637Z","shell.execute_reply":"2023-05-26T19:59:26.625456Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#Applied normalization for image\nim = imgCLAHE\nim = (im/(im.max()-im.min()))*255\nim_r = im[:,:,0]\nim_g = im[:,:,1]\nim_b = im[:,:,2]\nim_ = ((im_r**2)+(im_g**2)+(im_b**2))**(1/2)\nr=im_r/im_\ng=im_g/im_\nb=im_b/im_\nimg__ = cv2.merge((r,g,b))\nplt.imshow(img__) ","metadata":{"execution":{"iopub.status.busy":"2023-05-26T19:59:54.754029Z","iopub.execute_input":"2023-05-26T19:59:54.754478Z","iopub.status.idle":"2023-05-26T19:59:55.248102Z","shell.execute_reply.started":"2023-05-26T19:59:54.754436Z","shell.execute_reply":"2023-05-26T19:59:55.247143Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# AND apply CLAHE ONLY GREEN CHANNEL\nclahe = cv2.createCLAHE(clipLimit=2, tileGridSize=(64,64))\n#clahe_img0 = clahe.apply(img[:,:,0])\nclahe_img1 = clahe.apply(img[:,:,1]) # Green Channel\n#clahe_img2 = clahe.apply(img[:,:,2])\n\nimg_result = cv2.merge((img[:,:,0],clahe_img1,img[:,:,2]))\nimg_result = cv2.cvtColor(img_result, cv2.COLOR_BGR2RGB)\nplt.imshow(img_result)","metadata":{"execution":{"iopub.status.busy":"2021-06-12T19:34:14.750972Z","iopub.execute_input":"2021-06-12T19:34:14.751346Z","iopub.status.idle":"2021-06-12T19:34:15.047199Z","shell.execute_reply.started":"2021-06-12T19:34:14.751312Z","shell.execute_reply":"2021-06-12T19:34:15.046294Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"imgCLAHE.shape","metadata":{"execution":{"iopub.status.busy":"2021-06-12T19:34:15.048272Z","iopub.execute_input":"2021-06-12T19:34:15.04856Z","iopub.status.idle":"2021-06-12T19:34:15.056233Z","shell.execute_reply.started":"2021-06-12T19:34:15.048533Z","shell.execute_reply":"2021-06-12T19:34:15.054874Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#Normalization Function\ndef inorm(im): \n    im = (im/(im.max()-im.min()))*255\n    im_r = im[:,:,0]\n    im_g = im[:,:,1]\n    im_b = im[:,:,2]\n    im_ = ((im_r**2)+(im_g**2)+(im_b**2))**(1/3)\n    r=im_r/im_\n    g=im_g/im_\n    b=im_b/im_\n    img__ = cv2.merge((r,g,b))\n    return img__\n\n#Color Density Fuction\ndef rgb_density(img): \n    img = inorm(img)\n    r = img[:,:,0]\n    g = img[:,:,1]\n    b = img[:,:,2]\n    s = img.sum()\n    d1 = r.sum()/s\n    d2 = g.sum()/s\n    d3 = b.sum()/s\n    return [d1,d2,d3]\n    \nrgb_density(imgNORM)\n    ","metadata":{"execution":{"iopub.status.busy":"2021-06-12T19:34:15.060372Z","iopub.execute_input":"2021-06-12T19:34:15.06069Z","iopub.status.idle":"2021-06-12T19:34:15.220667Z","shell.execute_reply.started":"2021-06-12T19:34:15.060662Z","shell.execute_reply":"2021-06-12T19:34:15.219648Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#Color intensity value in CHALE applied image, respectively (R,G,B)\nrgb_density(imgCLAHE)","metadata":{"execution":{"iopub.status.busy":"2021-06-12T19:34:15.222138Z","iopub.execute_input":"2021-06-12T19:34:15.222485Z","iopub.status.idle":"2021-06-12T19:34:15.367694Z","shell.execute_reply.started":"2021-06-12T19:34:15.222453Z","shell.execute_reply":"2021-06-12T19:34:15.366603Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Color intensity value in the image with CHALE applied only to the green channel (R,G,B)\nrgb_density(img_result) ","metadata":{"execution":{"iopub.status.busy":"2021-06-12T19:34:15.369063Z","iopub.execute_input":"2021-06-12T19:34:15.369372Z","iopub.status.idle":"2021-06-12T19:34:15.519735Z","shell.execute_reply.started":"2021-06-12T19:34:15.369342Z","shell.execute_reply":"2021-06-12T19:34:15.518689Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"rgb_density(img__)","metadata":{"execution":{"iopub.status.busy":"2021-06-12T19:34:15.521145Z","iopub.execute_input":"2021-06-12T19:34:15.521638Z","iopub.status.idle":"2021-06-12T19:34:15.670828Z","shell.execute_reply.started":"2021-06-12T19:34:15.52159Z","shell.execute_reply":"2021-06-12T19:34:15.669466Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def summary(df, test_dir):\n    testfiles = os.listdir(test_dir)\n    df[['y_r']] = 0\n    df[['y_g']] = 0\n    df[['y_b']] = 0\n\n    for i in range(len(testfiles)):\n        img = cv2.imread(test_dir+testfile[i])\n        lab = cv2.cvtColor(img, cv2.COLOR_BGR2LAB)\n        l = lab[:,:,0]\n        a = lab[:,:,1]\n        b = lab[:,:,2]\n        clahe = cv2.createCLAHE(clipLimit=2, tileGridSize=(16,16))\n        clahe_img = clahe.apply(l)\n        img_result = cv2.merge((clahe_img,a,b))\n        imgCLAHE = cv2.cvtColor(img_result, cv2.COLOR_LAB2RGB)\n        yogunluk = rgb_density(imgCLAHE)\n        df['y_r'][i:i+1] = yogunluk[0]\n        df['y_g'][i:i+1] = yogunluk[1]\n        df['y_b'][i:i+1] = yogunluk[2]\n\n    return df","metadata":{"execution":{"iopub.status.busy":"2021-06-12T19:34:15.672514Z","iopub.execute_input":"2021-06-12T19:34:15.672977Z","iopub.status.idle":"2021-06-12T19:34:15.684201Z","shell.execute_reply.started":"2021-06-12T19:34:15.672919Z","shell.execute_reply":"2021-06-12T19:34:15.682803Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df_ = summary(df, test_dir)\ndf_.to_csv('result.csv')\ndf_","metadata":{"execution":{"iopub.status.busy":"2021-06-12T19:35:28.894015Z","iopub.execute_input":"2021-06-12T19:35:28.894506Z","iopub.status.idle":"2021-06-12T19:35:30.748084Z","shell.execute_reply.started":"2021-06-12T19:35:28.894471Z","shell.execute_reply":"2021-06-12T19:35:30.747072Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"YARARLANILAN KAYNAKLAR  \n[1] https://docs.opencv.org/master/d5/daf/tutorial_py_histogram_equalization.html  \n[2] https://medium.com/@ahmetkumas1/histogram-equalization-clahe-69cc9f83670c  \n\nAlso:  \nhttps://github.com/semskurto/wheat\n","metadata":{}}]}