Road Accident Detection And Avoidance Are A More Difficult And Challenging Problem In India As Poor Quality Of Construction Materials Get Used In Road Drainage System Construction. Due To The Above Problems, Roads Get Damaged Early And Potholes Appear On The Roads Which Cause Accidents. According To A Report Submitted By The Ministry Of Road Transport And Highways Transport Research Wing New Delhi In 2017, Approximately 4,64,910 Accidents Happen Per Year In India. This Paper Proposed A Deep Learning-based Model That Can Detect Potholes Early Using Images And Videos Which Can Reduce The Chances Of An Accident. This Model Is Basically Based On Transfer Learning, Faster Region-based Convolutional Neural Network(F-RCNN) And Inception-V2. There Are Many Models For Pothole Detection That Uses The Accelerometer (without Using Images And Videos) With Machine Learning Techniques, But A Less Number Of Pothole Detection Models Can Be Found Which Uses Only Machine Learning Techniques To Detect Potholes. The Results Of This Work Have Shown That Our Proposed Model Outperforms Other Existing Techniques Of Potholes Detection.
Digital Image Processing Domain Is Growing Day-by-day By Introducing Novel Technologies To Provide Assistance For Several Applications Such As Robotic Activities, Underwater Network Formation, And So On. In Particular, Underwater Image Processing Is Considered As The Crucial Task In Image Processing Industry Due To The Flow Of Light Waves That Are Not In The Specific And Expected Range Under The Water Level. While Image Restoration Technology Can Adequately Consider Removing This Same Haze From Source Images, They Need To Obtain Several Images From A Certain Place That Prevent It From Being Used In A Real-time System. To Overcome This Issue, A Deep Study Approach Is Developed By Providing Excellent Outcomes Of Deep Learning Approaches In Several Other Image Analysis Concerns Such As Colorizing Images Or Object Identification. A Convolution Neural Network (CNN) Model Is Trained To De-haze The Individual Images With Image Restoration In Order To Perform Further With An Image Improvement. The Proposed Approach Can Produce Images With Image Restoration Quality By Including A Standard Image Input And Here, The Neural Network Is Evaluated By Using Images And Features, Which Are Obtained From Separate Areas To Prove Its Capacity To Generalize. The Efficiency Of The Proposed Approach Is High When Compared To Other Existing Methods.
One Of The Important And Tedious Task In Agricultural Practices Is The Detection Of The Disease On Crops. It Requires Huge Time As Well As Skilled Labor. This Paper Proposes A Smart And Efficient Technique For Detection Of Crop Disease Which Uses Computer Vision And Machine Learning Techniques. The Proposed System Is Able To Detect 20 Different Diseases Of 5 Common Plants With 93% Accuracy.
Dogs Are One Of The Most Common Domestic Animals. Due To A Large Number Of Dogs, There Are Several Issues Such As Population Control, Decrease Outbreak Such As Rabies, Vaccination Control, And Legal Ownership. At Present, There Are Over 180 Dog Breeds. Each Dog Breed Has Specific Characteristics And Health Conditions. In Order To Provide Appropriate Treatments And Training, It Is Essential To Identify Individuals And Their Breeds. The Paper Presents The Classification Methods For Dog Breed Classification Using Two Image Processing Approaches 1) Conventional Based Approaches By Local Binary Pattern (LBP) And Histogram Of Oriented Gradient (HOG) 2) The Deep Learning Based Approach By Using Convolutional Neural Networks (CNN) With Transfer Learning. The Result Shows That Our Retrained CNN Model Performs Better In Classifying A Dog Breeds. It Achieves 96.75% Accuracy Compared With 79.25% Using The HOG Descriptor.
Detection Of Diseases In Plants Is A Significant Task That Has To Be Done In Agriculture. This Is Something On Which The Economy Profoundly Depends. Infection Discovery In Plants Is A Significant Job In The Agribusiness Field, As Having Diseases In Plants Is Very Common. To Recognize The Diseases In Leaves, A Continuous Observation Of The Plants Is Required. This Observation Or Continuous Monitoring Of The Plants Takes A Lot Of Human Effort And It Is Time-consuming Too. To Make It Simply Some Sort Of Programmed Strategy Is Required To Observe The Plants. Program Based Identification Of Diseases In Plants Makes Easier To Detect The Damaged Leaves And Reduces Human Efforts And Time-saving. The Proposed Algorithm Distinguishing Sickness In Plants And Classify Them More Accurately As Compared To Existing Techniques.
Historical Photographs Record The True Face Of A Moment In The Development Of Human History, They Have Authenticity, Vividness, And Unique Values. However, Due To Various Factors, Aging And Damage Will Occur. With The Development Of Computer Technology, The Restoration Technology Is More Used In Photo Restoration And Virtual Restoration Of Cultural Relics. This Paper First Analyzes The Principle Of Repairing Photo Archives Based On Computer Technology, And Then Uses The Combination Of Statistics And Computer Image Processing Technology To Detect And Repair The Scratches In Historical Photographs. And The Paper Establishes A Model Repair Framework, Which Provides A New Idea For The Repair Of Such Historical Photos. The Experimental Results Show That The Method Has A Significant Repair Effect.
One Of The Most Rapidly Spreading Cancers Among Various Other Types Of Cancers Known To Humans Is Skin Cancer. Melanoma Is The Worst And The Most Dangerous Type Of Skin Cancer That Appears Usually On The Skin Surface And Then Extends Deeper Into The Layers Of Skin. However, If Diagnosed At An Early Stage; The Survival Rate Of Melanoma Patients Is 96% With Simple And Economical Treatments. The Conventional Method Of Diagnosing Melanoma Involves Expert Dermatologists, Equipment, And Biopsies. To Avoid The Expensive Diagnosis, And To Assist Dermatologists, The Field Of Machine Learning Has Proven To Provide State Of The Art Solutions For Skin Cancer Detection At An Earlier Stage With High Accuracy. In This Paper, A Method For Skin Lesion Classification And Segmentation As Benign Or Malignant Is Proposed Using Image Processing And Machine Learning. A Novel Method Of Contrast Stretching Of Dermoscopic Images Based On The Methods Of Mean Values And Standard Deviation Of Pixels Is Proposed. Then The OTSU Thresholding Algorithm Is Applied For Image Segmentation. After The Segmentation, Features Including Gray Level Co-occurrence Matrix (GLCM) Features For Texture Identification, The Histogram Of Oriented Gradients (HOG) Object, And Color Identification Features Are Extracted From The Segmented Images. Principal Component Analysis (PCA) Reduction Of HOG Features Is Performed For Dimensionality Reduction. Synthetic Minority Oversampling Technique (SMOTE) Sampling Is Performed To Deal With The Class Imbalance Problem. The Feature Vector Is Then Standardized And Scaled. A Novel Approach Of Feature Selection Based On The Wrapper Method Is Proposed Before Classification. Classifiers Including Quadratic Discriminant, SVM (Medium Gaussian), And Random Forest Are Used For Classification. The Proposed Approach Is Verified On The Publicly Accessible Dataset Of ISIC-ISBI 2016. Maximum Accuracy Is Achieved Using The Random Forest Classifier. The Classification Accuracy Of The Proposed System With The Random Forest Classifier On ISIC-ISBI 2016 Is 93.89%. The Proposed Approach Of Contrast Stretching Before The Segmentation Gives Satisfactory Results Of Segmentation. Further, The Proposed Wrapper-based Approach Of Feature Selection In Combination With The Random Forest Classifier Gives Promising Results As Compared To Other Commonly Used Classifiers.
Over The Last Decades, The Incidence Of Skin Cancer, Melanoma And Non-melanoma, Has Increased At A Continuous Rate. In Particular For Melanoma, The Deadliest Type Of Skin Cancer, Early Detection Is Important To Increase Patient Prognosis. Recently, Deep Neural Networks (DNNs) Have Become Viable To Deal With Skin Cancer Detection. In This Work, We Present A Smartphone-based Application To Assist On Skin Cancer Detection. This Application Is Based On A Convolutional Neural Network(CNN) Trained On Clinical Images And Patients Demographics, Both Collected From Smartphones. Also, As Skin Cancer Datasets Are Imbalanced, We Present An Approach, Based On The Mutation Operator Of Differential Evolution (DE) Algorithm, To Balance Data. In This Sense, Beyond Provides A Flexible Tool To Assist Doctors On Skin Cancer Screening Phase, The Method Obtains Promising Results With A Balanced Accuracy Of 85% And A Recall Of 96%.