近年来,世界各国大力发展航空航天事业,卫星图像的目标检测在各行各业的应用得到了快速的发展,特别是军事侦查、海洋船舶和渔业管理等领域。由于卫星图像中有价值的信息极少,卫星图像数据规模巨大,这迫切需要智能辅助工具帮助相关从业人员从卫星图像中高效获取精确直观的信息。
本文利用深度学习技术，基于Yolov5算法框架实现卫星图像目标检测问题。

2 实现效果

实现效果如下：可以看出对船只、飞机等识别效果还是很好的。

3 Yolov5算法

简介
下图所示为 YOLOv5 的网络结构图，分为输入端，Backbone，Neck 和 Prediction 四个部分。其中，
输入端包括 Mosaic 数据增强、自适应图片缩放、自适应锚框计算，Backbone 包括 Focus 结构、CSP
结构，Neck 包括 FPN+PAN 结构，Prediction 包括GIOU_Loss 结构。

相关代码

class Yolo(object):
    def __init__(self, weights_file, verbose=True):
        self.verbose = verbose
        # detection params
        self.S = 7  # cell size
        self.B = 2  # boxes_per_cell
        self.classes = ["aeroplane", "bicycle", "bird", "boat", "bottle",
                        "bus", "car", "cat", "chair", "cow", "diningtable",
                        "dog", "horse", "motorbike", "person", "pottedplant",
                        "sheep", "sofa", "train","tvmonitor"]
        self.C = len(self.classes) # number of classes
        # offset for box center (top left point of each cell)
        self.x_offset = np.transpose(np.reshape(np.array([np.arange(self.S)]*self.S*self.B),
                                              [self.B, self.S, self.S]), [1, 2, 0])
        self.y_offset = np.transpose(self.x_offset, [1, 0, 2])

        self.threshold = 0.2  # confidence scores threhold
        self.iou_threshold = 0.4
        #  the maximum number of boxes to be selected by non max suppression
        self.max_output_size = 10

        self.sess = tf.Session()
        self._build_net()
        self._build_detector()
        self._load_weights(weights_file)

4 数据处理和训练

数据集
本项目使用 DOTA 数据集，原数据集中待检测的目标如下

原数据集中的标签如下

图像分割和尺寸调整
YOLO 模型的图像输入尺寸是固定的，由于原数据集中的图像尺寸不一，我们将原数据集中的图像按目标分布的位置分割成一个个包含目标的子图，并将每个子图尺寸调整为 1024×1024。分割前后的图像如所示。
分割前

分割后

模型训练
在 yolov5/ 目录，运行 train.py 文件开始训练：

python train.py --weight weights/yolov5s.pt --batch 16 --epochs 100 --cache

其中的参数说明：

weight：使用的预训练权重，这里示范使用的是 yolov5s 模型的预训练权重

batch：mini-batch 的大小，这里使用 16

epochs：训练的迭代次数，这里我们训练 100 个 epoch

cache：使用数据缓存，加速训练进程

相关代码

#部分代码
def train(hyp, opt, device, tb_writer=None):
    logger.info(f'Hyperparameters {hyp}')
    log_dir = Path(tb_writer.log_dir) if tb_writer else Path(opt.logdir) / 'evolve'  # logging directory
    wdir = log_dir / 'weights'  # weights directory
    os.makedirs(wdir, exist_ok=True)
    last = wdir / 'last.pt'
    best = wdir / 'best.pt'
    results_file = str(log_dir / 'results.txt')
    epochs, batch_size, total_batch_size, weights, rank = \
        opt.epochs, opt.batch_size, opt.total_batch_size, opt.weights, opt.global_rank

    # Save run settings
    with open(log_dir / 'hyp.yaml', 'w') as f:
        yaml.dump(hyp, f, sort_keys=False)
    with open(log_dir / 'opt.yaml', 'w') as f:
        yaml.dump(vars(opt), f, sort_keys=False)

    # Configure
    cuda = device.type != 'cpu'
    init_seeds(2 + rank)
    with open(opt.data) as f:
        data_dict = yaml.load(f, Loader=yaml.FullLoader)  # data dict
    with torch_distributed_zero_first(rank):
        check_dataset(data_dict)  # check
    train_path = data_dict['train']
    test_path = data_dict['val']
    nc, names = (1, ['item']) if opt.single_cls else (int(data_dict['nc']), data_dict['names'])  # number classes, names
    assert len(names) == nc, '%g names found for nc=%g dataset in %s' % (len(names), nc, opt.data)  # check

    # Model
    pretrained = weights.endswith('.pt')
    if pretrained:
        with torch_distributed_zero_first(rank):
            attempt_download(weights)  # download if not found locally
        ckpt = torch.load(weights, map_location=device)  # load checkpoint
        if 'anchors' in hyp and hyp['anchors']:
            ckpt['model'].yaml['anchors'] = round(hyp['anchors'])  # force autoanchor
        model = Model(opt.cfg or ckpt['model'].yaml, ch=3, nc=nc).to(device)  # create
        exclude = ['anchor'] if opt.cfg else []  # exclude keys
        state_dict = ckpt['model'].float().state_dict()  # to FP32
        state_dict = intersect_dicts(state_dict, model.state_dict(), exclude=exclude)  # intersect
        model.load_state_dict(state_dict, strict=False)  # load
        logger.info('Transferred %g/%g items from %s' % (len(state_dict), len(model.state_dict()), weights))  # report
    else:
        model = Model(opt.cfg, ch=3, nc=nc).to(device)  # create

    # Freeze
    freeze = ['', ]  # parameter names to freeze (full or partial)
    if any(freeze):
        for k, v in model.named_parameters():
            if any(x in k for x in freeze):
                print('freezing %s' % k)
                v.requires_grad = False

    # Optimizer
    nbs = 64  # nominal batch size
    accumulate = max(round(nbs / total_batch_size), 1)  # accumulate loss before optimizing
    hyp['weight_decay'] *= total_batch_size * accumulate / nbs  # scale weight_decay

    pg0, pg1, pg2 = [], [], []  # optimizer parameter groups
    for k, v in model.named_parameters():
        v.requires_grad = True
        if '.bias' in k:
            pg2.append(v)  # biases
        elif '.weight' in k and '.bn' not in k:
            pg1.append(v)  # apply weight decay
        else:
            pg0.append(v)  # all else

训练开始时的日志信息